1 Time Series 7-Day Forecasting with long short-term memory (LSTM) and embedding networks

1.1 Contents

• 1 Time series 7-day forecasting with LSTM
  • 1.1 Contents
  • 1.2 Imports
  • 1.3 Functions
    • 1.3.1 Function: feature_list
    • 1.3.2 Function: ebd_dim
    • 1.3.3 Function: split_sequences
    • 1.3.4 Function: to_embed
    • 1.3.5 Function: build_LSTM with embeddings
    • 1.3.6 Function: ndarray_to_input_list
    • 1.3.7 Function: ginic
    • 1.3.8 Function: gini_normalizedc
    • 1.3.9 Functions: rmse_y_y_pred
    • 1.3.10 Functions: plot_actual_predicted
  • 1.4 Load data
  • 1.5 Preprocessing
    • 1.5.1 Preprocessing: get feature and target for model training and testing with cross-validation
    • 1.5.2 Preprocessing: scale feature and target variables
  • 1.6 Train and test the LSTM model
    • 1.6.1 Train and test the LSTM model: train and test with cross-validation
    • 1.6.2 Summary and plot the model
  • 1.7 Organize and evaluate model performance
    • 1.7.1 Organize and evaluate model performance: train set
    • 1.7.2 Organize and evaluate model performance: validation set
    • 1.7.3 Organize and evaluate model performance: test set
    • 1.7.4 Organize and evaluate model performance: save
    • 1.7.5 Organize and evaluate model performance: plot

1.2 Imports

from google.colab import drive # (in Colab)
drive.mount('/content/drive')

Mounted at /content/drive

# Version control
!pip install pandas==1.3.2 
!pip install scipy==1.6.2
!pip install scikit-learn==0.24.2

Collecting pandas==1.3.2
  Downloading pandas-1.3.2-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.3 MB)
     |████████████████████████████████| 11.3 MB 8.2 MB/s 
Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.7/dist-packages (from pandas==1.3.2) (1.19.5)
Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas==1.3.2) (2018.9)
Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas==1.3.2) (2.8.2)
Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas==1.3.2) (1.15.0)
Installing collected packages: pandas
  Attempting uninstall: pandas
    Found existing installation: pandas 1.1.5
    Uninstalling pandas-1.1.5:
      Successfully uninstalled pandas-1.1.5
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
google-colab 1.0.0 requires pandas~=1.1.0; python_version >= "3.0", but you have pandas 1.3.2 which is incompatible.
Successfully installed pandas-1.3.2

Collecting scipy==1.6.2
  Downloading scipy-1.6.2-cp37-cp37m-manylinux1_x86_64.whl (27.4 MB)
     |████████████████████████████████| 27.4 MB 114 kB/s 
Requirement already satisfied: numpy<1.23.0,>=1.16.5 in /usr/local/lib/python3.7/dist-packages (from scipy==1.6.2) (1.19.5)
Installing collected packages: scipy
  Attempting uninstall: scipy
    Found existing installation: scipy 1.4.1
    Uninstalling scipy-1.4.1:
      Successfully uninstalled scipy-1.4.1
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
albumentations 0.1.12 requires imgaug<0.2.7,>=0.2.5, but you have imgaug 0.2.9 which is incompatible.
Successfully installed scipy-1.6.2
Collecting scikit-learn==0.24.2
  Downloading scikit_learn-0.24.2-cp37-cp37m-manylinux2010_x86_64.whl (22.3 MB)
     |████████████████████████████████| 22.3 MB 1.2 MB/s 
Requirement already satisfied: scipy>=0.19.1 in /usr/local/lib/python3.7/dist-packages (from scikit-learn==0.24.2) (1.6.2)
Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn==0.24.2) (1.0.1)
Requirement already satisfied: numpy>=1.13.3 in /usr/local/lib/python3.7/dist-packages (from scikit-learn==0.24.2) (1.19.5)
Collecting threadpoolctl>=2.0.0
  Downloading threadpoolctl-3.0.0-py3-none-any.whl (14 kB)
Installing collected packages: threadpoolctl, scikit-learn
  Attempting uninstall: scikit-learn
    Found existing installation: scikit-learn 0.22.2.post1
    Uninstalling scikit-learn-0.22.2.post1:
      Successfully uninstalled scikit-learn-0.22.2.post1
Successfully installed scikit-learn-0.24.2 threadpoolctl-3.0.0

import os
import pandas as pd
import numpy as np
import numpy.matlib
import matplotlib.pyplot as plt
import pickle
from pathlib import Path
from datetime import date, timedelta
import math
import warnings; warnings.simplefilter('ignore')

from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import MinMaxScaler
from sklearn.model_selection import StratifiedKFold
from sklearn.model_selection import TimeSeriesSplit
from sklearn.metrics import mean_squared_error

from keras.models import Model
from keras.layers import Input, Dense, Concatenate, Reshape, Dropout, LSTM, TimeDistributed
from keras.layers.embeddings import Embedding
from keras.utils.vis_utils import plot_model
from keras.callbacks import EarlyStopping

import tensorflow as tf
%load_ext tensorboard
import datetime, os

from contextlib import redirect_stdout
%matplotlib inline

# Set Matplotlib defaults
plt.style.use("seaborn-whitegrid")
plt.rc("figure", autolayout=True, figsize=(11, 5))
plt.rc(
    "axes",
    labelweight="bold",
    labelsize=14,
    titleweight="bold",
    titlesize=16,
    titlepad=10,
)
plot_params = dict(
    color="0.75",
    style=".-",
    markeredgecolor="0.25",
    markerfacecolor="0.25",
)

1.3 Functions

1.3.1 Functions: feature_list

# organize features in each row into 1) static categorical, 2) temporal categorical, 3) temporal continuous
def feature_list(country_id, row): 
    # Static Categorical
    country = country_id #0 country id
    # Temporal Categorical (datetime variables)
    dt = row[0].to_pydatetime()
    year = dt.year #1
    month = dt.month #2
    day = dt.day #3
    week_of_year = dt.isocalendar()[1] #4
    day_of_week = row[1].dayow #5
    holiday = row[1].holiday #6 holiday
    # Temporal Continuous (mobility variables-this will be lagged for LSTM)
    # Temporal Continuous (weather variables)
    cloudcover = float(row[1].cloudcover) #13 weather; cloudcover
    tempC = float(row[1].tempC) #14 weather; temparature
    humidity = float(row[1].humidity) #15 weather; humidity
    precipMM = float(row[1].precipMM) #16 weather; precipitation
    # Temporal Continuous (vaccination-this will be lagged for LSTM)           
    return [country], \
[year, month, day, week_of_year, day_of_week, holiday], \
[cloudcover, tempC, humidity, precipMM] # Static Categorical, Temporal Categorical, Temporal Continuous

1.3.2 Functions: ebd_dim

# determine the embedding dimensions
def ebd_dim(cat_dim):
    return min(600, round(1.6 * cat_dim ** .56))

1.3.3 Functions: split_sequences

# get the input and output sequences from the entire time series
def split_sequences(sequences, timestamp, n_steps_in, n_steps_out): 
    timestamps = sequences.index
    df_time0 = timestamps[0]
    df_time_end = timestamps[-1]
    dt_steps_in = timedelta(days=n_steps_in)
    dt_steps_out = timedelta(days=n_steps_out-1)   
    dt_1 = timedelta(days=1)
    if (timestamp-dt_steps_in>=df_time0) & (timestamp+dt_steps_out<=df_time_end): # if within bounds
        # gather input and output parts of the pattern 
        seq_x = sequences[timestamp-dt_steps_in:timestamp-dt_1] # input sequence (e.g. previous 14 days) 
        seq_y = sequences[timestamp:timestamp+dt_steps_out] # output sequence (e.g. next 7 days including the current timestamp)
    return list(seq_x), list(seq_y)

1.3.4 Functions: to_embed

# preprocess the embedding columns
def to_embed(cat_columns): 
    
    cat_conv_raw = []
    cat_conv_array = np.empty((cat_columns.shape[0],cat_columns.shape[1]))
    
    for c in range(cat_columns.shape[1]):
        cat_conv_raw.append(list(cat_columns[:,c]))
        raw_vals = np.unique(cat_columns[:,c])
        val_map = {}
        for i in range(len(raw_vals)):
            val_map[raw_vals[i]] = i
        cat_conv_array[:,c] = [val_map[j] for j in cat_conv_raw[c]]
    return cat_conv_array

1.3.5 Functions: build_LSTM_with_embeddings

def build_LSTM_with_embeddings(): 
    
    inputs = []
    embeddings = []
    input_dims = []
    
    input_stat_cat_00_country = Input(shape=(1,)) # country (0)
    inputs.append(input_stat_cat_00_country)
    input_dims.append(input_stat_cat_00_country.shape[-1])
    embedding = Embedding(23, 9, input_length=1)(input_stat_cat_00_country)
    embedding = Reshape(target_shape=(9,))(embedding)
    embeddings.append(embedding)
    
    input_temp_cat_01_year = Input(shape=(1,)) # year (1)
    inputs.append(input_temp_cat_01_year)
    input_dims.append(input_temp_cat_01_year.shape[-1])
    embedding = Embedding(2, 1, input_length=1)(input_temp_cat_01_year)
    embedding = Reshape(target_shape=(1,))(embedding)
    embeddings.append(embedding)
     
    input_temp_cat_02_month = Input(shape=(1,)) # month (2)
    inputs.append(input_temp_cat_02_month)
    input_dims.append(input_temp_cat_02_month.shape[-1])
    embedding = Embedding(12, 6, input_length=1)(input_temp_cat_02_month)
    embedding = Reshape(target_shape=(6,))(embedding)
    embeddings.append(embedding)    
    
    input_temp_cat_03_day = Input(shape=(1,)) # day (3)
    inputs.append(input_temp_cat_03_day)
    input_dims.append(input_temp_cat_03_day.shape[-1])
    embedding = Embedding(31, 11, input_length=1)(input_temp_cat_03_day)
    embedding = Reshape(target_shape=(11,))(embedding)
    embeddings.append(embedding)    
    
    input_temp_cat_04_week_of_year = Input(shape=(1,)) # week of year (4)
    inputs.append(input_temp_cat_04_week_of_year)
    input_dims.append(input_temp_cat_04_week_of_year.shape[-1])
    embedding = Embedding(53, 15, input_length=1)(input_temp_cat_04_week_of_year)
    embedding = Reshape(target_shape=(15,))(embedding)
    embeddings.append(embedding)
    
    input_temp_cat_05_day_of_week = Input(shape=(1,)) # day of week (5)
    inputs.append(input_temp_cat_05_day_of_week)
    input_dims.append(input_temp_cat_05_day_of_week.shape[-1])
    embedding = Embedding(7, 5, input_length=1)(input_temp_cat_05_day_of_week)
    embedding = Reshape(target_shape=(5,))(embedding)
    embeddings.append(embedding) 
    
    input_temp_cat_06_holiday = Input(shape=(1,)) # holiday (6)
    inputs.append(input_temp_cat_06_holiday)
    input_dims.append(input_temp_cat_06_holiday.shape[-1])
    embedding = Embedding(2, 2, input_length=1)(input_temp_cat_06_holiday)
    embedding = Reshape(target_shape=(2,))(embedding)
    embeddings.append(embedding)    
    
    input_weather = Input(shape=(4,)) # weather (temporal continuous) features without lagging (e.g. 4)
    inputs.append(input_weather)
    input_dims.append(input_weather.shape[-1])
    dense_weather = Dense(4)(input_weather)
    embeddings.append(dense_weather) 
    
    input_lstm = Input(shape=(14,8)) # input dim: batch size(omitted)-by-timesteps(14)-by-features(8; 6 mobility, vac, case) 
    inputs.append(input_lstm)
    input_dims.append(input_lstm.shape[-2]*input_lstm.shape[-1])
    
    x = Concatenate()(embeddings)
    x = Dense(90, activation='relu')(x)
    x = Dropout(.35)(x)
    x = Dense(40, activation='relu')(x)
    x = Dropout(.15)(x)
    x = Dense(14, activation='relu')(x)
    x = Dropout(.15)(x)
    
    lstm_ = LSTM(32, return_sequences=True, recurrent_dropout=0.1)(input_lstm) # output dim: (batch size-by-timesteps-by-hidden_size) 
    lstm_ = LSTM(64, return_sequences=True, recurrent_dropout=0.1)(lstm_) # output dim: (batch size-by-timesteps-by-hidden_size)
    lstm_ = TimeDistributed(Dense(16, activation='relu'))(lstm_)
    lstm_ = TimeDistributed(Dense(1, activation='relu'))(lstm_)
    lstm_ = Reshape((14,), input_shape=(None, 14, 1))(lstm_)
    
    x_lstm_ = Concatenate()([x, lstm_])
    x_lstm_ = Dense(14, activation='relu')(x_lstm_) 
    
    output = Dense(7, activation='linear')(x_lstm_)
    
    model = Model(inputs, output)
    
    model.compile(loss='mse', optimizer='adam', metrics=['mae'])
    
    return model, input_dims, inputs
# plot_model(model, to_file='lstm_embeddings_plot.png', show_shapes=True, show_layer_names=True)
# print(model.summary())

1.3.6 Functions: ndarray_to_input_list

def ndarray_to_input_list(ndarr,input_dims,keras_input_list): 
    input_list = []
    input_dim_cumsum = list(np.cumsum(input_dims))
    input_dim_cumsum.insert(0,0)
    curr_input_0 = input_dim_cumsum[:-1]
    for i,input_0 in enumerate(curr_input_0):
        if len(keras_input_list[i].shape)==2: # if input is 2d (batch-by-feature), no reshaping is needed
            input_list.append(ndarr[:,input_0:input_0+input_dims[i]])
        elif len(keras_input_list[i].shape)==3: # if input is 3d (batch-by-timestep-by-feature), reshaping is needed 
            to_reshape = ndarr[:,input_0:input_0+input_dims[i]]
            n_timesteps = keras_input_list[i].shape[-2]
            n_features = keras_input_list[i].shape[-1]
            reshaped = np.empty((ndarr.shape[0], n_timesteps, n_features))
            for i in range(n_features):
                reshaped[:, :n_timesteps, i] = to_reshape[:,i*n_timesteps:(i*n_timesteps+n_timesteps)]
            input_list.append(reshaped)        
    return input_list

1.3.7 Functions: ginic

#gini scoring function from kernel at: 
#https://www.kaggle.com/tezdhar/faster-gini-calculation
def ginic(actual, pred):
    pred_arr = np.asarray(pred)
    if len(pred_arr.shape)==2: # matrix 
        pred_mean = pred.copy()
    elif len(pred_arr.shape)==3: # tuple 
        pred_mean = np.nanmean(pred_arr,axis=0) # average across cross-validation folds
    n, c = actual.shape[0], actual.shape[1]
    gini_sum = []
    for col in range(c):
        a_s = actual[np.argsort(pred_mean[:,col]),col] # returns the indices that would sort an array
        a_c = a_s.cumsum()
        gini_sum.append((a_c.sum() / a_c[-1] - (n + 1) / 2.0)/n) 
    return gini_sum

1.3.8 Functions: gini_normalizedc

def gini_normalizedc(a, p):
    gini_a_p = ginic(a, p)
    gini_a_a = ginic(a, a)
    gini_norm = [gini_a_p[i]/gini_a_a[i] for i in range(len(gini_a_p))]
    return gini_norm

1.3.9 Functions: rmse_y_y_pred

def rmse_y_y_pred(rez_dict, n_steps_out):
    rmse_dict = {}
    countries = sorted(set(rez_dict['country']))
    for country in countries:
        country_idx = [cc==country for cc in rez_dict['country']]
        country_rmse = []
        for d in range(n_steps_out):
            country_rmse.append(mean_squared_error(rez_dict['y'][country_idx,d], rez_dict['y_pred'][country_idx,d], squared=False))
        rmse_dict[country] = country_rmse
    return rmse_dict 

1.3.10 Functions: plot_actual_predicted

def plot_actual_predicted(rez_dict, dict_country, save_name, n_steps_out=7):
    countries = sorted(set(rez_dict['country']))
    # plot for each of the 7-day forecast
    for country in countries:
        fig, ax = plt.subplots(n_steps_out, 1, figsize=(12, 36))
        fig.subplots_adjust(wspace=0.5, hspace=0.7)
        
        country_idx = [cc==country for cc in rez_dict['country']]
        timestamps = [i for i,v in zip(rez_dict['timestamp'],country_idx) if v]
        
        holidays = np.array(dict_country[country].loc[timestamps, 'holiday']==1) # holidays
        sundays = np.array(dict_country[country].loc[timestamps, 'dayow']==6) # Sundays
        holi_sun = np.logical_or(holidays, sundays)
        holiday_timestamps = [date.strftime(i, '%b-%d') for i,v in zip(rez_dict['timestamp'],holi_sun) if v]
        
        y_actual = rez_dict['y'][country_idx,:]
        y_pred = rez_dict['y_pred'][country_idx,:]
        
        for i in range(n_steps_out):
            ts_td = [t+timedelta(days=i) for t in timestamps]
            ts = list(map(lambda x:date.strftime(x,'%y-%b-%d'),ts_td))   
            
            if save_name == 'train':
                interval = 28
            elif save_name == 'val':
                interval = 14
            elif save_name == 'test':
                interval = 2
            
            ts_td_interval = [ts for i, ts in enumerate(ts_td) if i in np.arange(0, len(ts_td), interval)]
            ts_interval = list(map(lambda x:date.strftime(x,'%y-%b-%d'),ts_td_interval))     
            
            ax[i].plot(ts, y_actual[:,i], 'o-')
            ax[i].plot(ts, y_pred[:,i], 'o-')
            ax[i].xaxis.set_ticks(ts_interval)
            ax[i].set_xlabel('Date', fontsize=14, fontweight='bold')
            #ax[i].set_xlim(ts[0], ts[-1])
            ax[i].set_ylabel('Cases per million', fontsize=14, fontweight='bold')          
            ax[i].set_title(country + '_step_#' + str(i), fontweight='bold', fontsize=20)
            ax[i].legend(['y_actual','y_pred'], prop=dict(weight='bold',size=12))
            for tick in ax[i].get_xticklabels():
                tick.set_rotation(45)
            ax[i].tick_params(axis='x', labelsize=12)
            ax[i].tick_params(axis='y', labelsize=12)
            #for holi in holiday_timestamps:
            #    ax[i].axvspan(holi, holi, color='red', alpha=0.3, linewidth=2)
        # figure save
        fig.savefig(os.path.join('/Users/parkj/Documents/pyDat/dataSet/covid19_forecasting/covid19_figures/LSTM', \
                                  country+'_'+save_name+'_LSTM_7d.pdf'), tranparent=True)
        #fig.savefig(os.path.join('/content/drive/My Drive/Colab_data/covid19/Figure', \
        #                         country+'_'+save_name+'_MLP_7d.pdf'), tranparent=True)

def split_sequence_features(df_country, ts_curr): 
    f_rtrc, _ = split_sequences(df_country['rtrc'], ts_curr, n_steps_in=14, n_steps_out=0)  
    f_grph, _ = split_sequences(df_country['grph'], ts_curr, n_steps_in=14, n_steps_out=0)
    f_prks, _ = split_sequences(df_country['prks'], ts_curr, n_steps_in=14, n_steps_out=0)   
    f_tran, _ = split_sequences(df_country['tran'], ts_curr, n_steps_in=14, n_steps_out=0)   
    f_work, _ = split_sequences(df_country['work'], ts_curr, n_steps_in=14, n_steps_out=0)
    f_resi, _ = split_sequences(df_country['resi'], ts_curr, n_steps_in=14, n_steps_out=0)
    f_vac, _ = split_sequences(df_country['vac'], ts_curr, n_steps_in=14, n_steps_out=0)
    f_case, t_case = split_sequences(df_country['case_mil'], ts_curr, n_steps_in=14, n_steps_out=7)    
    return f_rtrc, f_grph, f_prks, f_tran, f_work, f_resi, f_vac, f_case, t_case

1.4 Load data

# load data from pickle file
filePath_pickle = Path('/Users/parkj/Documents/pyDat/dataSet/covid_country_data.pickle')
with open(filePath_pickle, 'rb') as f:
    dict_country = pickle.load(f)
# countries = ['AR', 'AT', 'AU', 'BE', 'CA', 'DE', 'DK', 'FI', 'FR', 'GB', 'ID', 'IE', 'IL', 'IN', 'IT', 'JP', 'KR', 'MX', 'NL', 'NO', 'RU', 'SG', 'US']

1.5 Preprocessing

1.5.1 Preprocessing: get feature and target for model training and testing with cross-validation

train_timestamp = []
train_country = []
train_stat_cat = []
train_temp_cat = []
train_temp_con = []
train_f_rtrc = []
train_f_grph = []
train_f_prks = []
train_f_tran = []
train_f_work = []
train_f_resi = []
train_f_vac = []
train_f_case = []
train_y_unscaled = []

test_timestamp = []
test_country = []
test_stat_cat = []
test_temp_cat = []
test_temp_con = []
test_f_rtrc = []
test_f_grph = []
test_f_prks = []
test_f_tran = []
test_f_work = []
test_f_resi = []
test_f_vac = []
test_f_case = []
test_y_unscaled = []

n_test = 21 # days
dt_test = timedelta(days=n_test-1)
n_steps_in = 14 # days (# previous cases)
dt_steps_in = timedelta(days=n_steps_in)
n_steps_out = 7 # days (# future cases to be predicted)
dt_steps_out = timedelta(days=n_steps_out-1)

for i, country_key in enumerate(dict_country.keys()):
    case_detection = 0
    df_country = dict_country[country_key]
    df_country.fillna(method='ffill',inplace=True) # forward fill NaNs
    df_time0 = df_country.index[0] # the first day of the data
    df_time_end = df_country.index[-1] # the last day of the data
    # split the df into train and test sets
    test_time0 = df_country.index[-1]-dt_test # the first date of test set 
    train_ind = df_country.index < test_time0 # training index
    # feature_list train 
    df_country_train = df_country.loc[train_ind] # train df
    for row in df_country_train.iterrows():
        ts_curr = row[0] 
        # case_mil lagging
        if (ts_curr-dt_steps_in>=df_time0) & (ts_curr+dt_steps_out<=df_time_end):
            # get feature and target variables
            f_rtrc, f_grph, f_prks, f_tran, f_work, f_resi, f_vac, f_case, t_case = split_sequence_features(df_country, ts_curr)     
            if (case_detection == 0) & (sum(f_case)>0): # to exclude days before 1st case detection
                case_detection = 1
            if case_detection == 1:
                fl_stat_cat, fl_temp_cat, fl_temp_con = feature_list(i, row) # get static categorical, temporal categorical, temporal continuous variables separately 
                # train data X (for embeddings)
                train_country.append(row[1].country_region_code)
                train_timestamp.append(ts_curr) # timestamps
                train_stat_cat.append(fl_stat_cat) # static categorical 
                train_temp_cat.append(fl_temp_cat) # temporal categorical 
                train_temp_con.append(fl_temp_con) # temporal continuous
                # train data X (for LSTM)
                train_f_rtrc.append(f_rtrc)
                train_f_grph.append(f_grph)
                train_f_prks.append(f_prks)
                train_f_tran.append(f_tran)
                train_f_work.append(f_work)
                train_f_resi.append(f_resi)
                train_f_vac.append(f_vac)
                train_f_case.append(f_case) # case_mil previous days to be used as features
                # train data y
                train_y_unscaled.append(t_case) # case_mil current & future days to be predicted
    # feature list test
    df_country_test = df_country.loc[~train_ind] # test df
    # feature list test 
    for row in df_country_test.iterrows():
        ts_curr = row[0]
        # case_mil lagging
        if (ts_curr-dt_steps_in>=df_time0) & (ts_curr+dt_steps_out<=df_time_end):        
            # get feature and target variables
            f_rtrc, f_grph, f_prks, f_tran, f_work, f_resi, f_vac, f_case, t_case = split_sequence_features(df_country, ts_curr)   
            fl_stat_cat, fl_temp_cat, fl_temp_con = feature_list(i, row) # get static categorical, temporal categorical, temporal continuous variables separately
            # test data X (for embeddings)
            test_country.append(row[1].country_region_code)
            test_timestamp.append(ts_curr)
            test_stat_cat.append(fl_stat_cat) # static categorical 
            test_temp_cat.append(fl_temp_cat) # temporal categorical 
            test_temp_con.append(fl_temp_con) # temporal continuous
            # test data X (for LSTM)
            test_f_rtrc.append(f_rtrc)
            test_f_grph.append(f_grph)
            test_f_prks.append(f_prks)
            test_f_tran.append(f_tran)
            test_f_work.append(f_work)
            test_f_resi.append(f_resi)
            test_f_vac.append(f_vac)      
            test_f_case.append(f_case) # case_mil previous days to be used as features
            # train data y
            test_y_unscaled.append(t_case) # case_mil current & future days to be predicted

1.5.2 Preprocessing: scale or embed feature and target variables

def matrix_scaler_over_all_columns(input_list, scaler_):
    concat_1d = []
    for m in input_list:
        concat_1d.append(np.reshape(m, (np.shape(m)[0]*np.shape(m)[1],1)))
    concat_1d_array = np.concatenate(concat_1d, axis=0)
    
    output_list = []
    for m in input_list:
        repmat = np.matlib.repmat(concat_1d_array, 1, np.shape(m)[1])
        #scaler_ = StandardScaler()
        scaler_.fit(repmat)
        output_list.append(scaler_.transform(m))       
    return output_list, scaler_

def matrix_scaler_each_column(input_list, scaler_):
    concat = np.concatenate(input_list, axis=0)
    scaler_.fit(concat)
    output_list = []
    for m in input_list:
        output_list.append(scaler_.transform(m))       
    return output_list

# embed static categorical 
train_stat_cat_embed = to_embed(np.array(train_stat_cat))
test_stat_cat_embed = to_embed(np.array(test_stat_cat))
# embed temporal categorical 
train_temp_cat_embed = to_embed(np.array(train_temp_cat))
test_temp_cat_embed = to_embed(np.array(test_temp_cat))
# scale temporal continuous
temp_con_scaled = matrix_scaler_each_column([train_temp_con, test_temp_con], MinMaxScaler())
train_temp_con_scaled = temp_con_scaled[0]
test_temp_con_scaled = temp_con_scaled[1]
# scale lagged temporal continuous (for LSTM)
# rtrc
rtrc_scaled, _ = matrix_scaler_over_all_columns([train_f_rtrc, test_f_rtrc], MinMaxScaler())
train_rtrc_scaled = rtrc_scaled[0]
test_rtrc_scaled = rtrc_scaled[1]
# grph
grph_scaled, _ = matrix_scaler_over_all_columns([train_f_grph, test_f_grph], MinMaxScaler())
train_grph_scaled = grph_scaled[0]
test_grph_scaled = grph_scaled[1]
# prks
prks_scaled, _ = matrix_scaler_over_all_columns([train_f_prks, test_f_prks], MinMaxScaler())
train_prks_scaled = prks_scaled[0]
test_prks_scaled = prks_scaled[1]
# tran
tran_scaled, _ = matrix_scaler_over_all_columns([train_f_tran, test_f_tran], MinMaxScaler())
train_tran_scaled = tran_scaled[0]
test_tran_scaled = tran_scaled[1]
# work
work_scaled, _ = matrix_scaler_over_all_columns([train_f_work, test_f_work], MinMaxScaler())
train_work_scaled = work_scaled[0]
test_work_scaled = work_scaled[1]
# resi
resi_scaled, _ = matrix_scaler_over_all_columns([train_f_resi, test_f_resi], MinMaxScaler())
train_resi_scaled = resi_scaled[0]
test_resi_scaled = resi_scaled[1]
# vac
vac_scaled, _ = matrix_scaler_over_all_columns([train_f_vac, test_f_vac], MinMaxScaler())
train_vac_scaled = vac_scaled[0]
test_vac_scaled = vac_scaled[1]
# case
case_scaled, scaler_y = matrix_scaler_over_all_columns([train_f_case, test_f_case, train_y_unscaled, test_y_unscaled], StandardScaler())
train_f_case_scaled = case_scaled[0]
test_f_case_scaled = case_scaled[1]
train_y = case_scaled[2]
test_y = case_scaled[3]

# concatenate features to get X
train_X = np.concatenate((train_stat_cat_embed, train_temp_cat_embed, train_temp_con_scaled, \
                          train_rtrc_scaled, train_grph_scaled, train_prks_scaled, train_tran_scaled, \
                          train_work_scaled, train_resi_scaled, train_vac_scaled, train_f_case_scaled), axis=1)

test_X = np.concatenate((test_stat_cat_embed, test_temp_cat_embed, test_temp_con_scaled, \
                         test_rtrc_scaled, test_grph_scaled, test_prks_scaled, test_tran_scaled, \
                         test_work_scaled, test_resi_scaled, test_vac_scaled, test_f_case_scaled), axis=1)
print("Number of train datapoints: ", len(train_y))
print("Number of test datapoints: ", len(test_y))

Number of train datapoints:  12541
Number of test datapoints:  345

1.6 Train and test the LSTM model with embeddings

1.6.0 Start tensorboard

logs_base_dir = "./logs"
os.makedirs(logs_base_dir, exist_ok=True)
%tensorboard --logdir {logs_base_dir}
tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=logs_base_dir, histogram_freq=1)

Output hidden; open in https://colab.research.google.com to view.

1.6.1 Train and test the LSTM model: train and test with cross-validation

# get the cross-validation folds for train and validation sets  
train_folds = []
validation_folds = []
# get the timestamps for train and validation folds
sorted_train_timestamp = sorted((set(train_timestamp))) # unique timestamps in the train set
tscv = TimeSeriesSplit(n_splits=5, test_size=21) # splitting train and validations sets for cross validation
for train_idx, validation_idx in tscv.split(sorted_train_timestamp): # get train and validation sets
    # print("TRAIN:", train_idx, "VALIDATION:", validation_idx)
    train_folds.append([sorted_train_timestamp[i] for i in train_idx]) # folds in train set
    validation_folds.append([sorted_train_timestamp[i] for i in validation_idx]) # folds in validation set

n_epochs = 1000
runs_per_fold = 5

cv_ginis = []
train_preds = []
train_idx_fold = [] 
val_preds = []
val_idx_fold = []
test_preds = []

# network training
for fold in range(len(train_folds)): # iterate cross-validation folds
    fold_idx_train = [ts in train_folds[fold] for ts in train_timestamp]
    train_idx_fold.append(fold_idx_train)
    fold_idx_val = [ts in validation_folds[fold] for ts in train_timestamp]
    val_idx_fold.append(fold_idx_val)
    X_train_f, X_val_f = train_X[fold_idx_train,:], train_X[fold_idx_val,:] # X_train for the current fold
    y_train_f, y_val_f = train_y[fold_idx_train,:], train_y[fold_idx_val,:] # y_train for the current fold
    
    train_preds_fold = []
    val_preds_fold = []
    
    for run_n in range(runs_per_fold): # iterate # of runs per fold
        # build the model
        NN, input_dim_NN, keras_input_list = build_LSTM_with_embeddings() # build the model and get input dimensions
        
        X_train_f_input = ndarray_to_input_list(X_train_f, input_dim_NN, keras_input_list)
        X_val_f_input = ndarray_to_input_list(X_val_f, input_dim_NN, keras_input_list)
        X_test_f_input = ndarray_to_input_list(test_X, input_dim_NN, keras_input_list)
        
        NN.fit(X_train_f_input, y_train_f, epochs=n_epochs, batch_size=2048, verbose=2, \
               callbacks = [EarlyStopping(monitor="loss", min_delta = 0.00001, patience = 50, mode = 'auto', restore_best_weights=True), tensorboard_callback])
   
        train_preds_fold.append(NN.predict(X_train_f_input))
        val_preds_fold.append(NN.predict(X_val_f_input))
        test_preds.append(NN.predict(X_test_f_input))
    
    cv_gini = gini_normalizedc(y_val_f, val_preds_fold) # actual, predicted
    cv_ginis.append(cv_gini)
    for i, cvg in enumerate(cv_gini):
        print('\nFold %d prediction cv gini_%d: %.5f\n' %(fold,i,cv_gini[i]))
    
    train_preds.append(np.mean(np.asarray(train_preds_fold),axis=0)) # ave
    val_preds.append(np.mean(np.asarray(val_preds_fold),axis=0))
    
cv_gini_mean = np.nanmean(np.asarray(cv_ginis),axis=0)
for i, g in enumerate(cv_gini_mean):
    print('Mean validation fold gini_%d: %.5f\n' %(i,g))

test_y_pred = np.nanmean(np.asarray(test_preds),axis=0) # average across cross-validation folds and runs

Streaming output truncated to the last 5000 lines.
6/6 - 0s - loss: 0.0837 - mae: 0.1327
Epoch 169/1000
6/6 - 0s - loss: 0.0815 - mae: 0.1306
Epoch 170/1000
6/6 - 0s - loss: 0.0820 - mae: 0.1306
Epoch 171/1000
6/6 - 0s - loss: 0.0820 - mae: 0.1310
Epoch 172/1000
6/6 - 0s - loss: 0.0816 - mae: 0.1314
Epoch 173/1000
6/6 - 0s - loss: 0.0831 - mae: 0.1319
Epoch 174/1000
6/6 - 1s - loss: 0.0821 - mae: 0.1319
Epoch 175/1000
6/6 - 0s - loss: 0.0816 - mae: 0.1304
Epoch 176/1000
6/6 - 0s - loss: 0.0818 - mae: 0.1296
Epoch 177/1000
6/6 - 0s - loss: 0.0814 - mae: 0.1296
Epoch 178/1000
6/6 - 0s - loss: 0.0809 - mae: 0.1295
Epoch 179/1000
6/6 - 0s - loss: 0.0804 - mae: 0.1303
Epoch 180/1000
6/6 - 0s - loss: 0.0816 - mae: 0.1296
Epoch 181/1000
6/6 - 0s - loss: 0.0805 - mae: 0.1300
Epoch 182/1000
6/6 - 0s - loss: 0.0808 - mae: 0.1297
Epoch 183/1000
6/6 - 0s - loss: 0.0800 - mae: 0.1287
Epoch 184/1000
6/6 - 0s - loss: 0.0793 - mae: 0.1287
Epoch 185/1000
6/6 - 0s - loss: 0.0800 - mae: 0.1290
Epoch 186/1000
6/6 - 1s - loss: 0.0796 - mae: 0.1288
Epoch 187/1000
6/6 - 0s - loss: 0.0801 - mae: 0.1296
Epoch 188/1000
6/6 - 0s - loss: 0.0798 - mae: 0.1282
Epoch 189/1000
6/6 - 0s - loss: 0.0800 - mae: 0.1288
Epoch 190/1000
6/6 - 0s - loss: 0.0801 - mae: 0.1287
Epoch 191/1000
6/6 - 0s - loss: 0.0789 - mae: 0.1284
Epoch 192/1000
6/6 - 0s - loss: 0.0791 - mae: 0.1278
Epoch 193/1000
6/6 - 1s - loss: 0.0796 - mae: 0.1279
Epoch 194/1000
6/6 - 0s - loss: 0.0794 - mae: 0.1275
Epoch 195/1000
6/6 - 0s - loss: 0.0791 - mae: 0.1285
Epoch 196/1000
6/6 - 0s - loss: 0.0780 - mae: 0.1270
Epoch 197/1000
6/6 - 0s - loss: 0.0781 - mae: 0.1267
Epoch 198/1000
6/6 - 0s - loss: 0.0782 - mae: 0.1269
Epoch 199/1000
6/6 - 0s - loss: 0.0793 - mae: 0.1274
Epoch 200/1000
6/6 - 0s - loss: 0.0780 - mae: 0.1261
Epoch 201/1000
6/6 - 0s - loss: 0.0786 - mae: 0.1264
Epoch 202/1000
6/6 - 0s - loss: 0.0784 - mae: 0.1268
Epoch 203/1000
6/6 - 0s - loss: 0.0780 - mae: 0.1263
Epoch 204/1000
6/6 - 0s - loss: 0.0771 - mae: 0.1262
Epoch 205/1000
6/6 - 0s - loss: 0.0773 - mae: 0.1267
Epoch 206/1000
6/6 - 1s - loss: 0.0783 - mae: 0.1265
Epoch 207/1000
6/6 - 0s - loss: 0.0767 - mae: 0.1257
Epoch 208/1000
6/6 - 1s - loss: 0.0772 - mae: 0.1253
Epoch 209/1000
6/6 - 0s - loss: 0.0776 - mae: 0.1266
Epoch 210/1000
6/6 - 1s - loss: 0.0766 - mae: 0.1254
Epoch 211/1000
6/6 - 0s - loss: 0.0770 - mae: 0.1269
Epoch 212/1000
6/6 - 0s - loss: 0.0763 - mae: 0.1246
Epoch 213/1000
6/6 - 0s - loss: 0.0763 - mae: 0.1248
Epoch 214/1000
6/6 - 0s - loss: 0.0785 - mae: 0.1257
Epoch 215/1000
6/6 - 0s - loss: 0.0763 - mae: 0.1251
Epoch 216/1000
6/6 - 0s - loss: 0.0768 - mae: 0.1252
Epoch 217/1000
6/6 - 0s - loss: 0.0760 - mae: 0.1250
Epoch 218/1000
6/6 - 0s - loss: 0.0759 - mae: 0.1236
Epoch 219/1000
6/6 - 0s - loss: 0.0758 - mae: 0.1250
Epoch 220/1000
6/6 - 0s - loss: 0.0761 - mae: 0.1246
Epoch 221/1000
6/6 - 0s - loss: 0.0763 - mae: 0.1241
Epoch 222/1000
6/6 - 1s - loss: 0.0757 - mae: 0.1246
Epoch 223/1000
6/6 - 1s - loss: 0.0758 - mae: 0.1237
Epoch 224/1000
6/6 - 0s - loss: 0.0756 - mae: 0.1244
Epoch 225/1000
6/6 - 1s - loss: 0.0750 - mae: 0.1238
Epoch 226/1000
6/6 - 0s - loss: 0.0750 - mae: 0.1243
Epoch 227/1000
6/6 - 0s - loss: 0.0739 - mae: 0.1232
Epoch 228/1000
6/6 - 0s - loss: 0.0745 - mae: 0.1243
Epoch 229/1000
6/6 - 1s - loss: 0.0747 - mae: 0.1241
Epoch 230/1000
6/6 - 0s - loss: 0.0756 - mae: 0.1233
Epoch 231/1000
6/6 - 0s - loss: 0.0742 - mae: 0.1232
Epoch 232/1000
6/6 - 0s - loss: 0.0748 - mae: 0.1224
Epoch 233/1000
6/6 - 0s - loss: 0.0750 - mae: 0.1228
Epoch 234/1000
6/6 - 0s - loss: 0.0747 - mae: 0.1232
Epoch 235/1000
6/6 - 0s - loss: 0.0735 - mae: 0.1226
Epoch 236/1000
6/6 - 0s - loss: 0.0736 - mae: 0.1226
Epoch 237/1000
6/6 - 0s - loss: 0.0742 - mae: 0.1228
Epoch 238/1000
6/6 - 0s - loss: 0.0733 - mae: 0.1224
Epoch 239/1000
6/6 - 0s - loss: 0.0738 - mae: 0.1224
Epoch 240/1000
6/6 - 0s - loss: 0.0729 - mae: 0.1215
Epoch 241/1000
6/6 - 1s - loss: 0.0742 - mae: 0.1220
Epoch 242/1000
6/6 - 0s - loss: 0.0741 - mae: 0.1232
Epoch 243/1000
6/6 - 0s - loss: 0.0733 - mae: 0.1238
Epoch 244/1000
6/6 - 0s - loss: 0.0739 - mae: 0.1226
Epoch 245/1000
6/6 - 0s - loss: 0.0738 - mae: 0.1225
Epoch 246/1000
6/6 - 1s - loss: 0.0737 - mae: 0.1217
Epoch 247/1000
6/6 - 0s - loss: 0.0736 - mae: 0.1218
Epoch 248/1000
6/6 - 0s - loss: 0.0730 - mae: 0.1225
Epoch 249/1000
6/6 - 0s - loss: 0.0732 - mae: 0.1215
Epoch 250/1000
6/6 - 0s - loss: 0.0733 - mae: 0.1214
Epoch 251/1000
6/6 - 0s - loss: 0.0733 - mae: 0.1230
Epoch 252/1000
6/6 - 0s - loss: 0.0738 - mae: 0.1218
Epoch 253/1000
6/6 - 0s - loss: 0.0738 - mae: 0.1213
Epoch 254/1000
6/6 - 0s - loss: 0.0725 - mae: 0.1208
Epoch 255/1000
6/6 - 1s - loss: 0.0726 - mae: 0.1208
Epoch 256/1000
6/6 - 1s - loss: 0.0729 - mae: 0.1209
Epoch 257/1000
6/6 - 0s - loss: 0.0722 - mae: 0.1198
Epoch 258/1000
6/6 - 0s - loss: 0.0727 - mae: 0.1214
Epoch 259/1000
6/6 - 0s - loss: 0.0735 - mae: 0.1207
Epoch 260/1000
6/6 - 1s - loss: 0.0725 - mae: 0.1211
Epoch 261/1000
6/6 - 0s - loss: 0.0715 - mae: 0.1197
Epoch 262/1000
6/6 - 1s - loss: 0.0726 - mae: 0.1197
Epoch 263/1000
6/6 - 0s - loss: 0.0719 - mae: 0.1202
Epoch 264/1000
6/6 - 1s - loss: 0.0721 - mae: 0.1212
Epoch 265/1000
6/6 - 0s - loss: 0.0717 - mae: 0.1214
Epoch 266/1000
6/6 - 1s - loss: 0.0710 - mae: 0.1202
Epoch 267/1000
6/6 - 0s - loss: 0.0720 - mae: 0.1203
Epoch 268/1000
6/6 - 0s - loss: 0.0705 - mae: 0.1192
Epoch 269/1000
6/6 - 1s - loss: 0.0712 - mae: 0.1202
Epoch 270/1000
6/6 - 1s - loss: 0.0717 - mae: 0.1207
Epoch 271/1000
6/6 - 1s - loss: 0.0709 - mae: 0.1204
Epoch 272/1000
6/6 - 1s - loss: 0.0716 - mae: 0.1195
Epoch 273/1000
6/6 - 0s - loss: 0.0707 - mae: 0.1189
Epoch 274/1000
6/6 - 0s - loss: 0.0708 - mae: 0.1186
Epoch 275/1000
6/6 - 0s - loss: 0.0705 - mae: 0.1190
Epoch 276/1000
6/6 - 0s - loss: 0.0715 - mae: 0.1199
Epoch 277/1000
6/6 - 1s - loss: 0.0692 - mae: 0.1192
Epoch 278/1000
6/6 - 0s - loss: 0.0722 - mae: 0.1206
Epoch 279/1000
6/6 - 0s - loss: 0.0733 - mae: 0.1242
Epoch 280/1000
6/6 - 0s - loss: 0.0718 - mae: 0.1226
Epoch 281/1000
6/6 - 0s - loss: 0.0720 - mae: 0.1229
Epoch 282/1000
6/6 - 0s - loss: 0.0710 - mae: 0.1205
Epoch 283/1000
6/6 - 0s - loss: 0.0712 - mae: 0.1210
Epoch 284/1000
6/6 - 0s - loss: 0.0710 - mae: 0.1203
Epoch 285/1000
6/6 - 0s - loss: 0.0717 - mae: 0.1212
Epoch 286/1000
6/6 - 0s - loss: 0.0700 - mae: 0.1191
Epoch 287/1000
6/6 - 0s - loss: 0.0708 - mae: 0.1206
Epoch 288/1000
6/6 - 0s - loss: 0.0693 - mae: 0.1182
Epoch 289/1000
6/6 - 0s - loss: 0.0697 - mae: 0.1187
Epoch 290/1000
6/6 - 0s - loss: 0.0700 - mae: 0.1178
Epoch 291/1000
6/6 - 1s - loss: 0.0694 - mae: 0.1178
Epoch 292/1000
6/6 - 0s - loss: 0.0693 - mae: 0.1176
Epoch 293/1000
6/6 - 0s - loss: 0.0685 - mae: 0.1169
Epoch 294/1000
6/6 - 0s - loss: 0.0687 - mae: 0.1180
Epoch 295/1000
6/6 - 0s - loss: 0.0692 - mae: 0.1173
Epoch 296/1000
6/6 - 0s - loss: 0.0690 - mae: 0.1176
Epoch 297/1000
6/6 - 0s - loss: 0.0691 - mae: 0.1174
Epoch 298/1000
6/6 - 0s - loss: 0.0701 - mae: 0.1183
Epoch 299/1000
6/6 - 0s - loss: 0.0696 - mae: 0.1181
Epoch 300/1000
6/6 - 0s - loss: 0.0684 - mae: 0.1181
Epoch 301/1000
6/6 - 0s - loss: 0.0686 - mae: 0.1176
Epoch 302/1000
6/6 - 0s - loss: 0.0680 - mae: 0.1178
Epoch 303/1000
6/6 - 0s - loss: 0.0686 - mae: 0.1179
Epoch 304/1000
6/6 - 0s - loss: 0.0679 - mae: 0.1174
Epoch 305/1000
6/6 - 0s - loss: 0.0671 - mae: 0.1169
Epoch 306/1000
6/6 - 0s - loss: 0.0677 - mae: 0.1160
Epoch 307/1000
6/6 - 0s - loss: 0.0696 - mae: 0.1172
Epoch 308/1000
6/6 - 0s - loss: 0.0686 - mae: 0.1175
Epoch 309/1000
6/6 - 1s - loss: 0.0673 - mae: 0.1170
Epoch 310/1000
6/6 - 1s - loss: 0.0677 - mae: 0.1163
Epoch 311/1000
6/6 - 0s - loss: 0.0694 - mae: 0.1181
Epoch 312/1000
6/6 - 0s - loss: 0.0682 - mae: 0.1164
Epoch 313/1000
6/6 - 0s - loss: 0.0686 - mae: 0.1172
Epoch 314/1000
6/6 - 0s - loss: 0.0683 - mae: 0.1176
Epoch 315/1000
6/6 - 0s - loss: 0.0677 - mae: 0.1172
Epoch 316/1000
6/6 - 0s - loss: 0.0671 - mae: 0.1154
Epoch 317/1000
6/6 - 1s - loss: 0.0674 - mae: 0.1157
Epoch 318/1000
6/6 - 0s - loss: 0.0671 - mae: 0.1152
Epoch 319/1000
6/6 - 0s - loss: 0.0673 - mae: 0.1151
Epoch 320/1000
6/6 - 0s - loss: 0.0678 - mae: 0.1155
Epoch 321/1000
6/6 - 0s - loss: 0.0666 - mae: 0.1159
Epoch 322/1000
6/6 - 0s - loss: 0.0675 - mae: 0.1150
Epoch 323/1000
6/6 - 0s - loss: 0.0670 - mae: 0.1153
Epoch 324/1000
6/6 - 0s - loss: 0.0669 - mae: 0.1166
Epoch 325/1000
6/6 - 1s - loss: 0.0679 - mae: 0.1158
Epoch 326/1000
6/6 - 0s - loss: 0.0651 - mae: 0.1155
Epoch 327/1000
6/6 - 0s - loss: 0.0653 - mae: 0.1152
Epoch 328/1000
6/6 - 1s - loss: 0.0654 - mae: 0.1146
Epoch 329/1000
6/6 - 0s - loss: 0.0661 - mae: 0.1151
Epoch 330/1000
6/6 - 0s - loss: 0.0653 - mae: 0.1139
Epoch 331/1000
6/6 - 0s - loss: 0.0662 - mae: 0.1149
Epoch 332/1000
6/6 - 0s - loss: 0.0658 - mae: 0.1146
Epoch 333/1000
6/6 - 1s - loss: 0.0649 - mae: 0.1142
Epoch 334/1000
6/6 - 0s - loss: 0.0661 - mae: 0.1142
Epoch 335/1000
6/6 - 0s - loss: 0.0664 - mae: 0.1155
Epoch 336/1000
6/6 - 0s - loss: 0.0663 - mae: 0.1142
Epoch 337/1000
6/6 - 0s - loss: 0.0648 - mae: 0.1142
Epoch 338/1000
6/6 - 0s - loss: 0.0648 - mae: 0.1138
Epoch 339/1000
6/6 - 0s - loss: 0.0658 - mae: 0.1150
Epoch 340/1000
6/6 - 0s - loss: 0.0655 - mae: 0.1141
Epoch 341/1000
6/6 - 0s - loss: 0.0647 - mae: 0.1136
Epoch 342/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1134
Epoch 343/1000
6/6 - 0s - loss: 0.0649 - mae: 0.1139
Epoch 344/1000
6/6 - 1s - loss: 0.0651 - mae: 0.1148
Epoch 345/1000
6/6 - 0s - loss: 0.0654 - mae: 0.1154
Epoch 346/1000
6/6 - 0s - loss: 0.0660 - mae: 0.1151
Epoch 347/1000
6/6 - 0s - loss: 0.0653 - mae: 0.1152
Epoch 348/1000
6/6 - 0s - loss: 0.0645 - mae: 0.1148
Epoch 349/1000
6/6 - 0s - loss: 0.0638 - mae: 0.1145
Epoch 350/1000
6/6 - 0s - loss: 0.0650 - mae: 0.1143
Epoch 351/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1139
Epoch 352/1000
6/6 - 0s - loss: 0.0653 - mae: 0.1155
Epoch 353/1000
6/6 - 1s - loss: 0.0653 - mae: 0.1146
Epoch 354/1000
6/6 - 0s - loss: 0.0647 - mae: 0.1136
Epoch 355/1000
6/6 - 0s - loss: 0.0645 - mae: 0.1135
Epoch 356/1000
6/6 - 0s - loss: 0.0641 - mae: 0.1136
Epoch 357/1000
6/6 - 0s - loss: 0.0639 - mae: 0.1133
Epoch 358/1000
6/6 - 1s - loss: 0.0641 - mae: 0.1126
Epoch 359/1000
6/6 - 0s - loss: 0.0629 - mae: 0.1125
Epoch 360/1000
6/6 - 0s - loss: 0.0630 - mae: 0.1126
Epoch 361/1000
6/6 - 0s - loss: 0.0640 - mae: 0.1127
Epoch 362/1000
6/6 - 0s - loss: 0.0633 - mae: 0.1126
Epoch 363/1000
6/6 - 0s - loss: 0.0638 - mae: 0.1124
Epoch 364/1000
6/6 - 0s - loss: 0.0622 - mae: 0.1122
Epoch 365/1000
6/6 - 0s - loss: 0.0626 - mae: 0.1121
Epoch 366/1000
6/6 - 0s - loss: 0.0630 - mae: 0.1127
Epoch 367/1000
6/6 - 0s - loss: 0.0626 - mae: 0.1118
Epoch 368/1000
6/6 - 0s - loss: 0.0621 - mae: 0.1122
Epoch 369/1000
6/6 - 0s - loss: 0.0627 - mae: 0.1124
Epoch 370/1000
6/6 - 0s - loss: 0.0632 - mae: 0.1138
Epoch 371/1000
6/6 - 0s - loss: 0.0620 - mae: 0.1119
Epoch 372/1000
6/6 - 0s - loss: 0.0619 - mae: 0.1116
Epoch 373/1000
6/6 - 0s - loss: 0.0620 - mae: 0.1126
Epoch 374/1000
6/6 - 0s - loss: 0.0620 - mae: 0.1118
Epoch 375/1000
6/6 - 0s - loss: 0.0616 - mae: 0.1133
Epoch 376/1000
6/6 - 0s - loss: 0.0630 - mae: 0.1137
Epoch 377/1000
6/6 - 0s - loss: 0.0621 - mae: 0.1125
Epoch 378/1000
6/6 - 0s - loss: 0.0625 - mae: 0.1134
Epoch 379/1000
6/6 - 0s - loss: 0.0623 - mae: 0.1139
Epoch 380/1000
6/6 - 0s - loss: 0.0644 - mae: 0.1159
Epoch 381/1000
6/6 - 0s - loss: 0.0631 - mae: 0.1161
Epoch 382/1000
6/6 - 0s - loss: 0.0620 - mae: 0.1131
Epoch 383/1000
6/6 - 0s - loss: 0.0614 - mae: 0.1118
Epoch 384/1000
6/6 - 1s - loss: 0.0616 - mae: 0.1120
Epoch 385/1000
6/6 - 0s - loss: 0.0614 - mae: 0.1126
Epoch 386/1000
6/6 - 1s - loss: 0.0624 - mae: 0.1120
Epoch 387/1000
6/6 - 1s - loss: 0.0614 - mae: 0.1114
Epoch 388/1000
6/6 - 0s - loss: 0.0607 - mae: 0.1106
Epoch 389/1000
6/6 - 0s - loss: 0.0619 - mae: 0.1113
Epoch 390/1000
6/6 - 0s - loss: 0.0605 - mae: 0.1105
Epoch 391/1000
6/6 - 0s - loss: 0.0613 - mae: 0.1115
Epoch 392/1000
6/6 - 0s - loss: 0.0615 - mae: 0.1109
Epoch 393/1000
6/6 - 0s - loss: 0.0603 - mae: 0.1106
Epoch 394/1000
6/6 - 0s - loss: 0.0610 - mae: 0.1111
Epoch 395/1000
6/6 - 0s - loss: 0.0611 - mae: 0.1109
Epoch 396/1000
6/6 - 0s - loss: 0.0610 - mae: 0.1120
Epoch 397/1000
6/6 - 0s - loss: 0.0607 - mae: 0.1114
Epoch 398/1000
6/6 - 0s - loss: 0.0607 - mae: 0.1105
Epoch 399/1000
6/6 - 0s - loss: 0.0599 - mae: 0.1101
Epoch 400/1000
6/6 - 1s - loss: 0.0599 - mae: 0.1118
Epoch 401/1000
6/6 - 1s - loss: 0.0598 - mae: 0.1122
Epoch 402/1000
6/6 - 1s - loss: 0.0601 - mae: 0.1112
Epoch 403/1000
6/6 - 1s - loss: 0.0601 - mae: 0.1101
Epoch 404/1000
6/6 - 0s - loss: 0.0589 - mae: 0.1100
Epoch 405/1000
6/6 - 0s - loss: 0.0600 - mae: 0.1104
Epoch 406/1000
6/6 - 1s - loss: 0.0598 - mae: 0.1101
Epoch 407/1000
6/6 - 0s - loss: 0.0591 - mae: 0.1107
Epoch 408/1000
6/6 - 0s - loss: 0.0596 - mae: 0.1106
Epoch 409/1000
6/6 - 0s - loss: 0.0608 - mae: 0.1121
Epoch 410/1000
6/6 - 0s - loss: 0.0590 - mae: 0.1106
Epoch 411/1000
6/6 - 0s - loss: 0.0593 - mae: 0.1111
Epoch 412/1000
6/6 - 0s - loss: 0.0596 - mae: 0.1099
Epoch 413/1000
6/6 - 0s - loss: 0.0586 - mae: 0.1096
Epoch 414/1000
6/6 - 0s - loss: 0.0591 - mae: 0.1092
Epoch 415/1000
6/6 - 0s - loss: 0.0592 - mae: 0.1089
Epoch 416/1000
6/6 - 0s - loss: 0.0588 - mae: 0.1096
Epoch 417/1000
6/6 - 0s - loss: 0.0581 - mae: 0.1093
Epoch 418/1000
6/6 - 1s - loss: 0.0585 - mae: 0.1090
Epoch 419/1000
6/6 - 0s - loss: 0.0584 - mae: 0.1102
Epoch 420/1000
6/6 - 0s - loss: 0.0584 - mae: 0.1116
Epoch 421/1000
6/6 - 0s - loss: 0.0581 - mae: 0.1110
Epoch 422/1000
6/6 - 1s - loss: 0.0574 - mae: 0.1095
Epoch 423/1000
6/6 - 0s - loss: 0.0580 - mae: 0.1095
Epoch 424/1000
6/6 - 0s - loss: 0.0585 - mae: 0.1097
Epoch 425/1000
6/6 - 1s - loss: 0.0570 - mae: 0.1093
Epoch 426/1000
6/6 - 0s - loss: 0.0576 - mae: 0.1102
Epoch 427/1000
6/6 - 0s - loss: 0.0579 - mae: 0.1090
Epoch 428/1000
6/6 - 1s - loss: 0.0579 - mae: 0.1094
Epoch 429/1000
6/6 - 1s - loss: 0.0571 - mae: 0.1085
Epoch 430/1000
6/6 - 0s - loss: 0.0578 - mae: 0.1090
Epoch 431/1000
6/6 - 0s - loss: 0.0569 - mae: 0.1091
Epoch 432/1000
6/6 - 0s - loss: 0.0567 - mae: 0.1101
Epoch 433/1000
6/6 - 0s - loss: 0.0576 - mae: 0.1101
Epoch 434/1000
6/6 - 0s - loss: 0.0571 - mae: 0.1096
Epoch 435/1000
6/6 - 0s - loss: 0.0573 - mae: 0.1097
Epoch 436/1000
6/6 - 0s - loss: 0.0573 - mae: 0.1091
Epoch 437/1000
6/6 - 1s - loss: 0.0561 - mae: 0.1078
Epoch 438/1000
6/6 - 0s - loss: 0.0568 - mae: 0.1082
Epoch 439/1000
6/6 - 0s - loss: 0.0562 - mae: 0.1081
Epoch 440/1000
6/6 - 0s - loss: 0.0569 - mae: 0.1088
Epoch 441/1000
6/6 - 0s - loss: 0.0569 - mae: 0.1093
Epoch 442/1000
6/6 - 0s - loss: 0.0563 - mae: 0.1089
Epoch 443/1000
6/6 - 0s - loss: 0.0559 - mae: 0.1086
Epoch 444/1000
6/6 - 0s - loss: 0.0566 - mae: 0.1080
Epoch 445/1000
6/6 - 0s - loss: 0.0567 - mae: 0.1086
Epoch 446/1000
6/6 - 0s - loss: 0.0555 - mae: 0.1079
Epoch 447/1000
6/6 - 0s - loss: 0.0558 - mae: 0.1082
Epoch 448/1000
6/6 - 1s - loss: 0.0558 - mae: 0.1090
Epoch 449/1000
6/6 - 0s - loss: 0.0563 - mae: 0.1087
Epoch 450/1000
6/6 - 0s - loss: 0.0552 - mae: 0.1076
Epoch 451/1000
6/6 - 0s - loss: 0.0560 - mae: 0.1084
Epoch 452/1000
6/6 - 0s - loss: 0.0553 - mae: 0.1074
Epoch 453/1000
6/6 - 0s - loss: 0.0548 - mae: 0.1069
Epoch 454/1000
6/6 - 0s - loss: 0.0552 - mae: 0.1076
Epoch 455/1000
6/6 - 0s - loss: 0.0559 - mae: 0.1076
Epoch 456/1000
6/6 - 0s - loss: 0.0553 - mae: 0.1077
Epoch 457/1000
6/6 - 0s - loss: 0.0557 - mae: 0.1083
Epoch 458/1000
6/6 - 0s - loss: 0.0539 - mae: 0.1079
Epoch 459/1000
6/6 - 0s - loss: 0.0551 - mae: 0.1076
Epoch 460/1000
6/6 - 0s - loss: 0.0536 - mae: 0.1067
Epoch 461/1000
6/6 - 0s - loss: 0.0542 - mae: 0.1072
Epoch 462/1000
6/6 - 1s - loss: 0.0558 - mae: 0.1081
Epoch 463/1000
6/6 - 0s - loss: 0.0553 - mae: 0.1096
Epoch 464/1000
6/6 - 0s - loss: 0.0549 - mae: 0.1091
Epoch 465/1000
6/6 - 0s - loss: 0.0546 - mae: 0.1091
Epoch 466/1000
6/6 - 0s - loss: 0.0549 - mae: 0.1090
Epoch 467/1000
6/6 - 1s - loss: 0.0549 - mae: 0.1094
Epoch 468/1000
6/6 - 0s - loss: 0.0540 - mae: 0.1084
Epoch 469/1000
6/6 - 0s - loss: 0.0548 - mae: 0.1097
Epoch 470/1000
6/6 - 0s - loss: 0.0546 - mae: 0.1082
Epoch 471/1000
6/6 - 0s - loss: 0.0543 - mae: 0.1074
Epoch 472/1000
6/6 - 0s - loss: 0.0536 - mae: 0.1071
Epoch 473/1000
6/6 - 0s - loss: 0.0538 - mae: 0.1069
Epoch 474/1000
6/6 - 0s - loss: 0.0532 - mae: 0.1070
Epoch 475/1000
6/6 - 0s - loss: 0.0542 - mae: 0.1071
Epoch 476/1000
6/6 - 0s - loss: 0.0552 - mae: 0.1074
Epoch 477/1000
6/6 - 0s - loss: 0.0531 - mae: 0.1066
Epoch 478/1000
6/6 - 0s - loss: 0.0537 - mae: 0.1069
Epoch 479/1000
6/6 - 0s - loss: 0.0530 - mae: 0.1078
Epoch 480/1000
6/6 - 1s - loss: 0.0523 - mae: 0.1064
Epoch 481/1000
6/6 - 0s - loss: 0.0530 - mae: 0.1069
Epoch 482/1000
6/6 - 0s - loss: 0.0532 - mae: 0.1066
Epoch 483/1000
6/6 - 1s - loss: 0.0529 - mae: 0.1069
Epoch 484/1000
6/6 - 0s - loss: 0.0524 - mae: 0.1064
Epoch 485/1000
6/6 - 0s - loss: 0.0530 - mae: 0.1072
Epoch 486/1000
6/6 - 0s - loss: 0.0530 - mae: 0.1068
Epoch 487/1000
6/6 - 0s - loss: 0.0521 - mae: 0.1077
Epoch 488/1000
6/6 - 0s - loss: 0.0522 - mae: 0.1060
Epoch 489/1000
6/6 - 0s - loss: 0.0531 - mae: 0.1064
Epoch 490/1000
6/6 - 0s - loss: 0.0524 - mae: 0.1056
Epoch 491/1000
6/6 - 0s - loss: 0.0522 - mae: 0.1060
Epoch 492/1000
6/6 - 0s - loss: 0.0524 - mae: 0.1062
Epoch 493/1000
6/6 - 0s - loss: 0.0533 - mae: 0.1069
Epoch 494/1000
6/6 - 0s - loss: 0.0533 - mae: 0.1064
Epoch 495/1000
6/6 - 0s - loss: 0.0527 - mae: 0.1063
Epoch 496/1000
6/6 - 0s - loss: 0.0520 - mae: 0.1064
Epoch 497/1000
6/6 - 1s - loss: 0.0519 - mae: 0.1060
Epoch 498/1000
6/6 - 0s - loss: 0.0516 - mae: 0.1074
Epoch 499/1000
6/6 - 0s - loss: 0.0522 - mae: 0.1070
Epoch 500/1000
6/6 - 1s - loss: 0.0519 - mae: 0.1070
Epoch 501/1000
6/6 - 0s - loss: 0.0522 - mae: 0.1064
Epoch 502/1000
6/6 - 0s - loss: 0.0516 - mae: 0.1055
Epoch 503/1000
6/6 - 0s - loss: 0.0514 - mae: 0.1053
Epoch 504/1000
6/6 - 0s - loss: 0.0519 - mae: 0.1055
Epoch 505/1000
6/6 - 0s - loss: 0.0522 - mae: 0.1054
Epoch 506/1000
6/6 - 0s - loss: 0.0508 - mae: 0.1055
Epoch 507/1000
6/6 - 0s - loss: 0.0509 - mae: 0.1056
Epoch 508/1000
6/6 - 0s - loss: 0.0513 - mae: 0.1057
Epoch 509/1000
6/6 - 0s - loss: 0.0507 - mae: 0.1044
Epoch 510/1000
6/6 - 0s - loss: 0.0513 - mae: 0.1056
Epoch 511/1000
6/6 - 0s - loss: 0.0507 - mae: 0.1061
Epoch 512/1000
6/6 - 0s - loss: 0.0507 - mae: 0.1049
Epoch 513/1000
6/6 - 0s - loss: 0.0508 - mae: 0.1050
Epoch 514/1000
6/6 - 0s - loss: 0.0502 - mae: 0.1053
Epoch 515/1000
6/6 - 0s - loss: 0.0511 - mae: 0.1049
Epoch 516/1000
6/6 - 0s - loss: 0.0498 - mae: 0.1049
Epoch 517/1000
6/6 - 0s - loss: 0.0506 - mae: 0.1055
Epoch 518/1000
6/6 - 0s - loss: 0.0500 - mae: 0.1052
Epoch 519/1000
6/6 - 0s - loss: 0.0508 - mae: 0.1057
Epoch 520/1000
6/6 - 0s - loss: 0.0507 - mae: 0.1054
Epoch 521/1000
6/6 - 0s - loss: 0.0504 - mae: 0.1045
Epoch 522/1000
6/6 - 1s - loss: 0.0516 - mae: 0.1073
Epoch 523/1000
6/6 - 1s - loss: 0.0495 - mae: 0.1060
Epoch 524/1000
6/6 - 0s - loss: 0.0499 - mae: 0.1043
Epoch 525/1000
6/6 - 0s - loss: 0.0502 - mae: 0.1054
Epoch 526/1000
6/6 - 1s - loss: 0.0504 - mae: 0.1051
Epoch 527/1000
6/6 - 1s - loss: 0.0495 - mae: 0.1047
Epoch 528/1000
6/6 - 0s - loss: 0.0497 - mae: 0.1044
Epoch 529/1000
6/6 - 0s - loss: 0.0503 - mae: 0.1049
Epoch 530/1000
6/6 - 0s - loss: 0.0502 - mae: 0.1042
Epoch 531/1000
6/6 - 0s - loss: 0.0494 - mae: 0.1047
Epoch 532/1000
6/6 - 1s - loss: 0.0496 - mae: 0.1046
Epoch 533/1000
6/6 - 0s - loss: 0.0499 - mae: 0.1052
Epoch 534/1000
6/6 - 0s - loss: 0.0484 - mae: 0.1045
Epoch 535/1000
6/6 - 0s - loss: 0.0504 - mae: 0.1063
Epoch 536/1000
6/6 - 1s - loss: 0.0493 - mae: 0.1055
Epoch 537/1000
6/6 - 0s - loss: 0.0500 - mae: 0.1055
Epoch 538/1000
6/6 - 0s - loss: 0.0499 - mae: 0.1050
Epoch 539/1000
6/6 - 0s - loss: 0.0490 - mae: 0.1046
Epoch 540/1000
6/6 - 0s - loss: 0.0494 - mae: 0.1045
Epoch 541/1000
6/6 - 0s - loss: 0.0486 - mae: 0.1038
Epoch 542/1000
6/6 - 0s - loss: 0.0491 - mae: 0.1037
Epoch 543/1000
6/6 - 1s - loss: 0.0488 - mae: 0.1043
Epoch 544/1000
6/6 - 0s - loss: 0.0494 - mae: 0.1051
Epoch 545/1000
6/6 - 0s - loss: 0.0490 - mae: 0.1043
Epoch 546/1000
6/6 - 1s - loss: 0.0486 - mae: 0.1041
Epoch 547/1000
6/6 - 0s - loss: 0.0493 - mae: 0.1040
Epoch 548/1000
6/6 - 0s - loss: 0.0488 - mae: 0.1040
Epoch 549/1000
6/6 - 0s - loss: 0.0491 - mae: 0.1043
Epoch 550/1000
6/6 - 0s - loss: 0.0486 - mae: 0.1039
Epoch 551/1000
6/6 - 1s - loss: 0.0482 - mae: 0.1034
Epoch 552/1000
6/6 - 0s - loss: 0.0488 - mae: 0.1041
Epoch 553/1000
6/6 - 1s - loss: 0.0484 - mae: 0.1045
Epoch 554/1000
6/6 - 1s - loss: 0.0486 - mae: 0.1037
Epoch 555/1000
6/6 - 0s - loss: 0.0484 - mae: 0.1036
Epoch 556/1000
6/6 - 0s - loss: 0.0487 - mae: 0.1051
Epoch 557/1000
6/6 - 0s - loss: 0.0489 - mae: 0.1053
Epoch 558/1000
6/6 - 0s - loss: 0.0482 - mae: 0.1042
Epoch 559/1000
6/6 - 0s - loss: 0.0475 - mae: 0.1033
Epoch 560/1000
6/6 - 0s - loss: 0.0473 - mae: 0.1033
Epoch 561/1000
6/6 - 0s - loss: 0.0481 - mae: 0.1035
Epoch 562/1000
6/6 - 0s - loss: 0.0478 - mae: 0.1035
Epoch 563/1000
6/6 - 0s - loss: 0.0471 - mae: 0.1031
Epoch 564/1000
6/6 - 0s - loss: 0.0474 - mae: 0.1032
Epoch 565/1000
6/6 - 0s - loss: 0.0478 - mae: 0.1037
Epoch 566/1000
6/6 - 0s - loss: 0.0480 - mae: 0.1038
Epoch 567/1000
6/6 - 0s - loss: 0.0469 - mae: 0.1031
Epoch 568/1000
6/6 - 0s - loss: 0.0481 - mae: 0.1033
Epoch 569/1000
6/6 - 0s - loss: 0.0479 - mae: 0.1040
Epoch 570/1000
6/6 - 0s - loss: 0.0480 - mae: 0.1045
Epoch 571/1000
6/6 - 0s - loss: 0.0470 - mae: 0.1039
Epoch 572/1000
6/6 - 0s - loss: 0.0474 - mae: 0.1039
Epoch 573/1000
6/6 - 0s - loss: 0.0476 - mae: 0.1033
Epoch 574/1000
6/6 - 0s - loss: 0.0472 - mae: 0.1033
Epoch 575/1000
6/6 - 0s - loss: 0.0473 - mae: 0.1041
Epoch 576/1000
6/6 - 1s - loss: 0.0474 - mae: 0.1037
Epoch 577/1000
6/6 - 1s - loss: 0.0474 - mae: 0.1036
Epoch 578/1000
6/6 - 0s - loss: 0.0465 - mae: 0.1024
Epoch 579/1000
6/6 - 0s - loss: 0.0473 - mae: 0.1037
Epoch 580/1000
6/6 - 0s - loss: 0.0470 - mae: 0.1034
Epoch 581/1000
6/6 - 0s - loss: 0.0466 - mae: 0.1028
Epoch 582/1000
6/6 - 0s - loss: 0.0468 - mae: 0.1029
Epoch 583/1000
6/6 - 0s - loss: 0.0469 - mae: 0.1030
Epoch 584/1000
6/6 - 0s - loss: 0.0462 - mae: 0.1029
Epoch 585/1000
6/6 - 0s - loss: 0.0463 - mae: 0.1027
Epoch 586/1000
6/6 - 0s - loss: 0.0467 - mae: 0.1029
Epoch 587/1000
6/6 - 0s - loss: 0.0464 - mae: 0.1029
Epoch 588/1000
6/6 - 0s - loss: 0.0475 - mae: 0.1031
Epoch 589/1000
6/6 - 0s - loss: 0.0467 - mae: 0.1035
Epoch 590/1000
6/6 - 0s - loss: 0.0469 - mae: 0.1034
Epoch 591/1000
6/6 - 0s - loss: 0.0474 - mae: 0.1031
Epoch 592/1000
6/6 - 0s - loss: 0.0465 - mae: 0.1042
Epoch 593/1000
6/6 - 0s - loss: 0.0471 - mae: 0.1048
Epoch 594/1000
6/6 - 1s - loss: 0.0459 - mae: 0.1033
Epoch 595/1000
6/6 - 0s - loss: 0.0460 - mae: 0.1032
Epoch 596/1000
6/6 - 0s - loss: 0.0463 - mae: 0.1028
Epoch 597/1000
6/6 - 1s - loss: 0.0467 - mae: 0.1035
Epoch 598/1000
6/6 - 0s - loss: 0.0459 - mae: 0.1026
Epoch 599/1000
6/6 - 0s - loss: 0.0454 - mae: 0.1016
Epoch 600/1000
6/6 - 0s - loss: 0.0459 - mae: 0.1024
Epoch 601/1000
6/6 - 0s - loss: 0.0457 - mae: 0.1017
Epoch 602/1000
6/6 - 0s - loss: 0.0454 - mae: 0.1015
Epoch 603/1000
6/6 - 0s - loss: 0.0459 - mae: 0.1028
Epoch 604/1000
6/6 - 0s - loss: 0.0458 - mae: 0.1030
Epoch 605/1000
6/6 - 0s - loss: 0.0458 - mae: 0.1024
Epoch 606/1000
6/6 - 0s - loss: 0.0461 - mae: 0.1028
Epoch 607/1000
6/6 - 0s - loss: 0.0460 - mae: 0.1038
Epoch 608/1000
6/6 - 0s - loss: 0.0459 - mae: 0.1024
Epoch 609/1000
6/6 - 1s - loss: 0.0463 - mae: 0.1029
Epoch 610/1000
6/6 - 0s - loss: 0.0456 - mae: 0.1032
Epoch 611/1000
6/6 - 0s - loss: 0.0456 - mae: 0.1020
Epoch 612/1000
6/6 - 0s - loss: 0.0450 - mae: 0.1017
Epoch 613/1000
6/6 - 0s - loss: 0.0452 - mae: 0.1018
Epoch 614/1000
6/6 - 0s - loss: 0.0454 - mae: 0.1026
Epoch 615/1000
6/6 - 0s - loss: 0.0449 - mae: 0.1021
Epoch 616/1000
6/6 - 0s - loss: 0.0454 - mae: 0.1019
Epoch 617/1000
6/6 - 0s - loss: 0.0451 - mae: 0.1028
Epoch 618/1000
6/6 - 1s - loss: 0.0451 - mae: 0.1034
Epoch 619/1000
6/6 - 0s - loss: 0.0444 - mae: 0.1017
Epoch 620/1000
6/6 - 0s - loss: 0.0445 - mae: 0.1011
Epoch 621/1000
6/6 - 0s - loss: 0.0452 - mae: 0.1020
Epoch 622/1000
6/6 - 0s - loss: 0.0444 - mae: 0.1013
Epoch 623/1000
6/6 - 1s - loss: 0.0449 - mae: 0.1018
Epoch 624/1000
6/6 - 0s - loss: 0.0452 - mae: 0.1017
Epoch 625/1000
6/6 - 1s - loss: 0.0452 - mae: 0.1025
Epoch 626/1000
6/6 - 0s - loss: 0.0453 - mae: 0.1025
Epoch 627/1000
6/6 - 0s - loss: 0.0450 - mae: 0.1013
Epoch 628/1000
6/6 - 0s - loss: 0.0445 - mae: 0.1009
Epoch 629/1000
6/6 - 0s - loss: 0.0444 - mae: 0.1009
Epoch 630/1000
6/6 - 0s - loss: 0.0442 - mae: 0.1011
Epoch 631/1000
6/6 - 0s - loss: 0.0443 - mae: 0.1009
Epoch 632/1000
6/6 - 0s - loss: 0.0445 - mae: 0.1031
Epoch 633/1000
6/6 - 0s - loss: 0.0439 - mae: 0.1011
Epoch 634/1000
6/6 - 0s - loss: 0.0441 - mae: 0.1015
Epoch 635/1000
6/6 - 0s - loss: 0.0441 - mae: 0.1019
Epoch 636/1000
6/6 - 0s - loss: 0.0446 - mae: 0.1017
Epoch 637/1000
6/6 - 1s - loss: 0.0447 - mae: 0.1017
Epoch 638/1000
6/6 - 0s - loss: 0.0439 - mae: 0.1012
Epoch 639/1000
6/6 - 1s - loss: 0.0439 - mae: 0.1017
Epoch 640/1000
6/6 - 1s - loss: 0.0444 - mae: 0.1016
Epoch 641/1000
6/6 - 0s - loss: 0.0444 - mae: 0.1012
Epoch 642/1000
6/6 - 1s - loss: 0.0437 - mae: 0.1006
Epoch 643/1000
6/6 - 1s - loss: 0.0437 - mae: 0.1007
Epoch 644/1000
6/6 - 0s - loss: 0.0441 - mae: 0.1019
Epoch 645/1000
6/6 - 0s - loss: 0.0438 - mae: 0.1018
Epoch 646/1000
6/6 - 0s - loss: 0.0437 - mae: 0.1012
Epoch 647/1000
6/6 - 0s - loss: 0.0432 - mae: 0.1007
Epoch 648/1000
6/6 - 0s - loss: 0.0442 - mae: 0.1020
Epoch 649/1000
6/6 - 0s - loss: 0.0437 - mae: 0.1020
Epoch 650/1000
6/6 - 0s - loss: 0.0431 - mae: 0.1008
Epoch 651/1000
6/6 - 0s - loss: 0.0433 - mae: 0.1006
Epoch 652/1000
6/6 - 0s - loss: 0.0433 - mae: 0.1006
Epoch 653/1000
6/6 - 0s - loss: 0.0431 - mae: 0.1013
Epoch 654/1000
6/6 - 0s - loss: 0.0430 - mae: 0.1007
Epoch 655/1000
6/6 - 0s - loss: 0.0432 - mae: 0.1008
Epoch 656/1000
6/6 - 0s - loss: 0.0441 - mae: 0.1007
Epoch 657/1000
6/6 - 1s - loss: 0.0427 - mae: 0.0998
Epoch 658/1000
6/6 - 0s - loss: 0.0426 - mae: 0.1002
Epoch 659/1000
6/6 - 0s - loss: 0.0440 - mae: 0.1014
Epoch 660/1000
6/6 - 0s - loss: 0.0438 - mae: 0.1014
Epoch 661/1000
6/6 - 0s - loss: 0.0431 - mae: 0.1003
Epoch 662/1000
6/6 - 0s - loss: 0.0436 - mae: 0.1007
Epoch 663/1000
6/6 - 0s - loss: 0.0433 - mae: 0.1011
Epoch 664/1000
6/6 - 0s - loss: 0.0430 - mae: 0.1007
Epoch 665/1000
6/6 - 0s - loss: 0.0437 - mae: 0.1006
Epoch 666/1000
6/6 - 0s - loss: 0.0427 - mae: 0.1008
Epoch 667/1000
6/6 - 0s - loss: 0.0424 - mae: 0.1001
Epoch 668/1000
6/6 - 0s - loss: 0.0430 - mae: 0.1001
Epoch 669/1000
6/6 - 0s - loss: 0.0431 - mae: 0.1003
Epoch 670/1000
6/6 - 0s - loss: 0.0427 - mae: 0.1002
Epoch 671/1000
6/6 - 0s - loss: 0.0424 - mae: 0.0999
Epoch 672/1000
6/6 - 0s - loss: 0.0422 - mae: 0.1001
Epoch 673/1000
6/6 - 0s - loss: 0.0420 - mae: 0.1001
Epoch 674/1000
6/6 - 1s - loss: 0.0430 - mae: 0.1002
Epoch 675/1000
6/6 - 0s - loss: 0.0429 - mae: 0.1002
Epoch 676/1000
6/6 - 0s - loss: 0.0423 - mae: 0.1007
Epoch 677/1000
6/6 - 0s - loss: 0.0427 - mae: 0.1005
Epoch 678/1000
6/6 - 0s - loss: 0.0426 - mae: 0.1013
Epoch 679/1000
6/6 - 0s - loss: 0.0427 - mae: 0.1011
Epoch 680/1000
6/6 - 0s - loss: 0.0423 - mae: 0.1001
Epoch 681/1000
6/6 - 0s - loss: 0.0419 - mae: 0.0995
Epoch 682/1000
6/6 - 0s - loss: 0.0420 - mae: 0.0998
Epoch 683/1000
6/6 - 0s - loss: 0.0420 - mae: 0.0998
Epoch 684/1000
6/6 - 0s - loss: 0.0421 - mae: 0.1003
Epoch 685/1000
6/6 - 1s - loss: 0.0415 - mae: 0.0993
Epoch 686/1000
6/6 - 0s - loss: 0.0419 - mae: 0.0988
Epoch 687/1000
6/6 - 0s - loss: 0.0419 - mae: 0.0996
Epoch 688/1000
6/6 - 0s - loss: 0.0412 - mae: 0.0992
Epoch 689/1000
6/6 - 0s - loss: 0.0408 - mae: 0.0989
Epoch 690/1000
6/6 - 0s - loss: 0.0416 - mae: 0.0992
Epoch 691/1000
6/6 - 0s - loss: 0.0416 - mae: 0.0992
Epoch 692/1000
6/6 - 0s - loss: 0.0416 - mae: 0.0990
Epoch 693/1000
6/6 - 1s - loss: 0.0418 - mae: 0.1003
Epoch 694/1000
6/6 - 0s - loss: 0.0418 - mae: 0.0996
Epoch 695/1000
6/6 - 0s - loss: 0.0421 - mae: 0.1013
Epoch 696/1000
6/6 - 0s - loss: 0.0417 - mae: 0.1004
Epoch 697/1000
6/6 - 0s - loss: 0.0416 - mae: 0.1000
Epoch 698/1000
6/6 - 0s - loss: 0.0412 - mae: 0.0992
Epoch 699/1000
6/6 - 0s - loss: 0.0409 - mae: 0.0991
Epoch 700/1000
6/6 - 1s - loss: 0.0408 - mae: 0.0988
Epoch 701/1000
6/6 - 0s - loss: 0.0410 - mae: 0.0983
Epoch 702/1000
6/6 - 0s - loss: 0.0410 - mae: 0.0991
Epoch 703/1000
6/6 - 0s - loss: 0.0410 - mae: 0.0995
Epoch 704/1000
6/6 - 0s - loss: 0.0409 - mae: 0.0990
Epoch 705/1000
6/6 - 0s - loss: 0.0402 - mae: 0.0987
Epoch 706/1000
6/6 - 0s - loss: 0.0408 - mae: 0.0990
Epoch 707/1000
6/6 - 0s - loss: 0.0408 - mae: 0.0994
Epoch 708/1000
6/6 - 0s - loss: 0.0404 - mae: 0.0986
Epoch 709/1000
6/6 - 0s - loss: 0.0406 - mae: 0.0989
Epoch 710/1000
6/6 - 0s - loss: 0.0401 - mae: 0.0985
Epoch 711/1000
6/6 - 0s - loss: 0.0407 - mae: 0.0988
Epoch 712/1000
6/6 - 0s - loss: 0.0405 - mae: 0.0990
Epoch 713/1000
6/6 - 0s - loss: 0.0402 - mae: 0.0989
Epoch 714/1000
6/6 - 0s - loss: 0.0405 - mae: 0.0986
Epoch 715/1000
6/6 - 0s - loss: 0.0402 - mae: 0.0982
Epoch 716/1000
6/6 - 0s - loss: 0.0400 - mae: 0.0977
Epoch 717/1000
6/6 - 0s - loss: 0.0400 - mae: 0.0980
Epoch 718/1000
6/6 - 0s - loss: 0.0394 - mae: 0.0979
Epoch 719/1000
6/6 - 0s - loss: 0.0402 - mae: 0.0979
Epoch 720/1000
6/6 - 0s - loss: 0.0404 - mae: 0.0989
Epoch 721/1000
6/6 - 0s - loss: 0.0404 - mae: 0.0990
Epoch 722/1000
6/6 - 0s - loss: 0.0404 - mae: 0.0987
Epoch 723/1000
6/6 - 0s - loss: 0.0406 - mae: 0.0998
Epoch 724/1000
6/6 - 1s - loss: 0.0399 - mae: 0.0983
Epoch 725/1000
6/6 - 0s - loss: 0.0402 - mae: 0.0990
Epoch 726/1000
6/6 - 0s - loss: 0.0400 - mae: 0.1000
Epoch 727/1000
6/6 - 0s - loss: 0.0404 - mae: 0.1001
Epoch 728/1000
6/6 - 0s - loss: 0.0405 - mae: 0.0995
Epoch 729/1000
6/6 - 0s - loss: 0.0407 - mae: 0.0992
Epoch 730/1000
6/6 - 0s - loss: 0.0401 - mae: 0.0995
Epoch 731/1000
6/6 - 0s - loss: 0.0399 - mae: 0.0993
Epoch 732/1000
6/6 - 0s - loss: 0.0396 - mae: 0.0989
Epoch 733/1000
6/6 - 0s - loss: 0.0394 - mae: 0.0983
Epoch 734/1000
6/6 - 0s - loss: 0.0389 - mae: 0.0972
Epoch 735/1000
6/6 - 0s - loss: 0.0393 - mae: 0.0983
Epoch 736/1000
6/6 - 1s - loss: 0.0394 - mae: 0.0985
Epoch 737/1000
6/6 - 0s - loss: 0.0394 - mae: 0.0980
Epoch 738/1000
6/6 - 0s - loss: 0.0388 - mae: 0.0985
Epoch 739/1000
6/6 - 1s - loss: 0.0397 - mae: 0.1003
Epoch 740/1000
6/6 - 0s - loss: 0.0390 - mae: 0.0980
Epoch 741/1000
6/6 - 0s - loss: 0.0389 - mae: 0.0976
Epoch 742/1000
6/6 - 0s - loss: 0.0389 - mae: 0.0986
Epoch 743/1000
6/6 - 0s - loss: 0.0391 - mae: 0.0986
Epoch 744/1000
6/6 - 0s - loss: 0.0390 - mae: 0.0987
Epoch 745/1000
6/6 - 0s - loss: 0.0383 - mae: 0.0977
Epoch 746/1000
6/6 - 1s - loss: 0.0384 - mae: 0.0968
Epoch 747/1000
6/6 - 0s - loss: 0.0384 - mae: 0.0972
Epoch 748/1000
6/6 - 0s - loss: 0.0388 - mae: 0.0977
Epoch 749/1000
6/6 - 0s - loss: 0.0388 - mae: 0.0975
Epoch 750/1000
6/6 - 0s - loss: 0.0384 - mae: 0.0972
Epoch 751/1000
6/6 - 0s - loss: 0.0388 - mae: 0.0978
Epoch 752/1000
6/6 - 0s - loss: 0.0383 - mae: 0.0972
Epoch 753/1000
6/6 - 0s - loss: 0.0385 - mae: 0.0972
Epoch 754/1000
6/6 - 0s - loss: 0.0386 - mae: 0.0976
Epoch 755/1000
6/6 - 0s - loss: 0.0388 - mae: 0.0979
Epoch 756/1000
6/6 - 0s - loss: 0.0380 - mae: 0.0969
Epoch 757/1000
6/6 - 0s - loss: 0.0382 - mae: 0.0961
Epoch 758/1000
6/6 - 0s - loss: 0.0386 - mae: 0.0972
Epoch 759/1000
6/6 - 0s - loss: 0.0381 - mae: 0.0976
Epoch 760/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0963
Epoch 761/1000
6/6 - 0s - loss: 0.0380 - mae: 0.0965
Epoch 762/1000
6/6 - 1s - loss: 0.0387 - mae: 0.0969
Epoch 763/1000
6/6 - 0s - loss: 0.0380 - mae: 0.0967
Epoch 764/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0965
Epoch 765/1000
6/6 - 0s - loss: 0.0378 - mae: 0.0965
Epoch 766/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0970
Epoch 767/1000
6/6 - 1s - loss: 0.0386 - mae: 0.0978
Epoch 768/1000
6/6 - 0s - loss: 0.0381 - mae: 0.0968
Epoch 769/1000
6/6 - 0s - loss: 0.0376 - mae: 0.0970
Epoch 770/1000
6/6 - 1s - loss: 0.0377 - mae: 0.0970
Epoch 771/1000
6/6 - 0s - loss: 0.0380 - mae: 0.0975
Epoch 772/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0981
Epoch 773/1000
6/6 - 1s - loss: 0.0381 - mae: 0.0984
Epoch 774/1000
6/6 - 0s - loss: 0.0375 - mae: 0.0965
Epoch 775/1000
6/6 - 0s - loss: 0.0370 - mae: 0.0957
Epoch 776/1000
6/6 - 0s - loss: 0.0371 - mae: 0.0961
Epoch 777/1000
6/6 - 0s - loss: 0.0374 - mae: 0.0961
Epoch 778/1000
6/6 - 0s - loss: 0.0367 - mae: 0.0956
Epoch 779/1000
6/6 - 0s - loss: 0.0369 - mae: 0.0961
Epoch 780/1000
6/6 - 0s - loss: 0.0372 - mae: 0.0958
Epoch 781/1000
6/6 - 0s - loss: 0.0367 - mae: 0.0957
Epoch 782/1000
6/6 - 1s - loss: 0.0369 - mae: 0.0959
Epoch 783/1000
6/6 - 0s - loss: 0.0373 - mae: 0.0960
Epoch 784/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0975
Epoch 785/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0987
Epoch 786/1000
6/6 - 0s - loss: 0.0369 - mae: 0.0968
Epoch 787/1000
6/6 - 0s - loss: 0.0372 - mae: 0.0969
Epoch 788/1000
6/6 - 0s - loss: 0.0370 - mae: 0.0965
Epoch 789/1000
6/6 - 0s - loss: 0.0371 - mae: 0.0964
Epoch 790/1000
6/6 - 0s - loss: 0.0371 - mae: 0.0978
Epoch 791/1000
6/6 - 0s - loss: 0.0373 - mae: 0.0967
Epoch 792/1000
6/6 - 0s - loss: 0.0368 - mae: 0.0971
Epoch 793/1000
6/6 - 0s - loss: 0.0369 - mae: 0.0969
Epoch 794/1000
6/6 - 0s - loss: 0.0363 - mae: 0.0966
Epoch 795/1000
6/6 - 0s - loss: 0.0365 - mae: 0.0960
Epoch 796/1000
6/6 - 0s - loss: 0.0367 - mae: 0.0960
Epoch 797/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0953
Epoch 798/1000
6/6 - 1s - loss: 0.0364 - mae: 0.0961
Epoch 799/1000
6/6 - 0s - loss: 0.0361 - mae: 0.0960
Epoch 800/1000
6/6 - 0s - loss: 0.0367 - mae: 0.0960
Epoch 801/1000
6/6 - 0s - loss: 0.0361 - mae: 0.0958
Epoch 802/1000
6/6 - 0s - loss: 0.0362 - mae: 0.0964
Epoch 803/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0959
Epoch 804/1000
6/6 - 0s - loss: 0.0357 - mae: 0.0959
Epoch 805/1000
6/6 - 0s - loss: 0.0361 - mae: 0.0952
Epoch 806/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0954
Epoch 807/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0956
Epoch 808/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0952
Epoch 809/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0952
Epoch 810/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0953
Epoch 811/1000
6/6 - 0s - loss: 0.0347 - mae: 0.0954
Epoch 812/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0951
Epoch 813/1000
6/6 - 1s - loss: 0.0357 - mae: 0.0957
Epoch 814/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0952
Epoch 815/1000
6/6 - 0s - loss: 0.0363 - mae: 0.0960
Epoch 816/1000
6/6 - 0s - loss: 0.0363 - mae: 0.0957
Epoch 817/1000
6/6 - 0s - loss: 0.0357 - mae: 0.0949
Epoch 818/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0959
Epoch 819/1000
6/6 - 0s - loss: 0.0355 - mae: 0.0964
Epoch 820/1000
6/6 - 0s - loss: 0.0355 - mae: 0.0944
Epoch 821/1000
6/6 - 0s - loss: 0.0350 - mae: 0.0956
Epoch 822/1000
6/6 - 1s - loss: 0.0353 - mae: 0.0962
Epoch 823/1000
6/6 - 0s - loss: 0.0351 - mae: 0.0950
Epoch 824/1000
6/6 - 0s - loss: 0.0350 - mae: 0.0949
Epoch 825/1000
6/6 - 0s - loss: 0.0349 - mae: 0.0952
Epoch 826/1000
6/6 - 0s - loss: 0.0351 - mae: 0.0945
Epoch 827/1000
6/6 - 0s - loss: 0.0353 - mae: 0.0947
Epoch 828/1000
6/6 - 0s - loss: 0.0350 - mae: 0.0941
Epoch 829/1000
6/6 - 0s - loss: 0.0350 - mae: 0.0948
Epoch 830/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0936
Epoch 831/1000
6/6 - 0s - loss: 0.0348 - mae: 0.0945
Epoch 832/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0938
Epoch 833/1000
6/6 - 0s - loss: 0.0344 - mae: 0.0940
Epoch 834/1000
6/6 - 1s - loss: 0.0344 - mae: 0.0938
Epoch 835/1000
6/6 - 1s - loss: 0.0344 - mae: 0.0942
Epoch 836/1000
6/6 - 0s - loss: 0.0348 - mae: 0.0946
Epoch 837/1000
6/6 - 1s - loss: 0.0347 - mae: 0.0955
Epoch 838/1000
6/6 - 0s - loss: 0.0344 - mae: 0.0948
Epoch 839/1000
6/6 - 0s - loss: 0.0345 - mae: 0.0944
Epoch 840/1000
6/6 - 0s - loss: 0.0345 - mae: 0.0942
Epoch 841/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0938
Epoch 842/1000
6/6 - 0s - loss: 0.0341 - mae: 0.0941
Epoch 843/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0934
Epoch 844/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0938
Epoch 845/1000
6/6 - 0s - loss: 0.0345 - mae: 0.0940
Epoch 846/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0943
Epoch 847/1000
6/6 - 0s - loss: 0.0348 - mae: 0.0936
Epoch 848/1000
6/6 - 0s - loss: 0.0336 - mae: 0.0934
Epoch 849/1000
6/6 - 0s - loss: 0.0337 - mae: 0.0935
Epoch 850/1000
6/6 - 0s - loss: 0.0336 - mae: 0.0934
Epoch 851/1000
6/6 - 0s - loss: 0.0339 - mae: 0.0931
Epoch 852/1000
6/6 - 0s - loss: 0.0346 - mae: 0.0941
Epoch 853/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0934
Epoch 854/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0944
Epoch 855/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0940
Epoch 856/1000
6/6 - 0s - loss: 0.0333 - mae: 0.0935
Epoch 857/1000
6/6 - 0s - loss: 0.0333 - mae: 0.0931
Epoch 858/1000
6/6 - 1s - loss: 0.0336 - mae: 0.0933
Epoch 859/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0938
Epoch 860/1000
6/6 - 0s - loss: 0.0341 - mae: 0.0937
Epoch 861/1000
6/6 - 0s - loss: 0.0342 - mae: 0.0944
Epoch 862/1000
6/6 - 0s - loss: 0.0339 - mae: 0.0941
Epoch 863/1000
6/6 - 0s - loss: 0.0333 - mae: 0.0941
Epoch 864/1000
6/6 - 0s - loss: 0.0339 - mae: 0.0935
Epoch 865/1000
6/6 - 0s - loss: 0.0338 - mae: 0.0940
Epoch 866/1000
6/6 - 0s - loss: 0.0341 - mae: 0.0949
Epoch 867/1000
6/6 - 0s - loss: 0.0337 - mae: 0.0935
Epoch 868/1000
6/6 - 0s - loss: 0.0336 - mae: 0.0935
Epoch 869/1000
6/6 - 0s - loss: 0.0334 - mae: 0.0936
Epoch 870/1000
6/6 - 0s - loss: 0.0333 - mae: 0.0936
Epoch 871/1000
6/6 - 0s - loss: 0.0337 - mae: 0.0947
Epoch 872/1000
6/6 - 0s - loss: 0.0336 - mae: 0.0939
Epoch 873/1000
6/6 - 0s - loss: 0.0330 - mae: 0.0927
Epoch 874/1000
6/6 - 0s - loss: 0.0331 - mae: 0.0929
Epoch 875/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0928
Epoch 876/1000
6/6 - 0s - loss: 0.0331 - mae: 0.0931
Epoch 877/1000
6/6 - 0s - loss: 0.0333 - mae: 0.0928
Epoch 878/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0931
Epoch 879/1000
6/6 - 0s - loss: 0.0328 - mae: 0.0930
Epoch 880/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0933
Epoch 881/1000
6/6 - 0s - loss: 0.0334 - mae: 0.0948
Epoch 882/1000
6/6 - 0s - loss: 0.0324 - mae: 0.0928
Epoch 883/1000
6/6 - 0s - loss: 0.0325 - mae: 0.0923
Epoch 884/1000
6/6 - 0s - loss: 0.0332 - mae: 0.0928
Epoch 885/1000
6/6 - 1s - loss: 0.0325 - mae: 0.0927
Epoch 886/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0923
Epoch 887/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0924
Epoch 888/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0931
Epoch 889/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0924
Epoch 890/1000
6/6 - 0s - loss: 0.0328 - mae: 0.0921
Epoch 891/1000
6/6 - 0s - loss: 0.0328 - mae: 0.0925
Epoch 892/1000
6/6 - 0s - loss: 0.0327 - mae: 0.0924
Epoch 893/1000
6/6 - 1s - loss: 0.0323 - mae: 0.0929
Epoch 894/1000
6/6 - 1s - loss: 0.0323 - mae: 0.0926
Epoch 895/1000
6/6 - 0s - loss: 0.0322 - mae: 0.0921
Epoch 896/1000
6/6 - 0s - loss: 0.0325 - mae: 0.0919
Epoch 897/1000
6/6 - 0s - loss: 0.0317 - mae: 0.0917
Epoch 898/1000
6/6 - 0s - loss: 0.0319 - mae: 0.0919
Epoch 899/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0918
Epoch 900/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0931
Epoch 901/1000
6/6 - 0s - loss: 0.0319 - mae: 0.0921
Epoch 902/1000
6/6 - 0s - loss: 0.0328 - mae: 0.0927
Epoch 903/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0929
Epoch 904/1000
6/6 - 0s - loss: 0.0321 - mae: 0.0917
Epoch 905/1000
6/6 - 0s - loss: 0.0321 - mae: 0.0914
Epoch 906/1000
6/6 - 0s - loss: 0.0319 - mae: 0.0920
Epoch 907/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0915
Epoch 908/1000
6/6 - 0s - loss: 0.0324 - mae: 0.0923
Epoch 909/1000
6/6 - 0s - loss: 0.0319 - mae: 0.0917
Epoch 910/1000
6/6 - 0s - loss: 0.0320 - mae: 0.0916
Epoch 911/1000
6/6 - 0s - loss: 0.0317 - mae: 0.0916
Epoch 912/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0909
Epoch 913/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0914
Epoch 914/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0914
Epoch 915/1000
6/6 - 0s - loss: 0.0316 - mae: 0.0914
Epoch 916/1000
6/6 - 0s - loss: 0.0311 - mae: 0.0912
Epoch 917/1000
6/6 - 0s - loss: 0.0311 - mae: 0.0913
Epoch 918/1000
6/6 - 0s - loss: 0.0312 - mae: 0.0911
Epoch 919/1000
6/6 - 0s - loss: 0.0313 - mae: 0.0908
Epoch 920/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0911
Epoch 921/1000
6/6 - 0s - loss: 0.0307 - mae: 0.0906
Epoch 922/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0915
Epoch 923/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0911
Epoch 924/1000
6/6 - 0s - loss: 0.0317 - mae: 0.0913
Epoch 925/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0913
Epoch 926/1000
6/6 - 0s - loss: 0.0316 - mae: 0.0915
Epoch 927/1000
6/6 - 1s - loss: 0.0310 - mae: 0.0911
Epoch 928/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0922
Epoch 929/1000
6/6 - 0s - loss: 0.0316 - mae: 0.0938
Epoch 930/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0916
Epoch 931/1000
6/6 - 0s - loss: 0.0312 - mae: 0.0909
Epoch 932/1000
6/6 - 0s - loss: 0.0315 - mae: 0.0914
Epoch 933/1000
6/6 - 0s - loss: 0.0309 - mae: 0.0908
Epoch 934/1000
6/6 - 0s - loss: 0.0308 - mae: 0.0909
Epoch 935/1000
6/6 - 0s - loss: 0.0308 - mae: 0.0909
Epoch 936/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0918
Epoch 937/1000
6/6 - 0s - loss: 0.0311 - mae: 0.0911
Epoch 938/1000
6/6 - 0s - loss: 0.0311 - mae: 0.0908
Epoch 939/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0910
Epoch 940/1000
6/6 - 0s - loss: 0.0307 - mae: 0.0909
Epoch 941/1000
6/6 - 0s - loss: 0.0313 - mae: 0.0925
Epoch 942/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0925
Epoch 943/1000
6/6 - 0s - loss: 0.0308 - mae: 0.0906
Epoch 944/1000
6/6 - 0s - loss: 0.0309 - mae: 0.0908
Epoch 945/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0916
Epoch 946/1000
6/6 - 0s - loss: 0.0306 - mae: 0.0911
Epoch 947/1000
6/6 - 0s - loss: 0.0309 - mae: 0.0910
Epoch 948/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0918
Epoch 949/1000
6/6 - 0s - loss: 0.0307 - mae: 0.0916
Epoch 950/1000
6/6 - 1s - loss: 0.0307 - mae: 0.0910
Epoch 951/1000
6/6 - 0s - loss: 0.0304 - mae: 0.0900
Epoch 952/1000
6/6 - 0s - loss: 0.0307 - mae: 0.0908
Epoch 953/1000
6/6 - 0s - loss: 0.0305 - mae: 0.0905
Epoch 954/1000
6/6 - 0s - loss: 0.0306 - mae: 0.0904
Epoch 955/1000
6/6 - 1s - loss: 0.0310 - mae: 0.0905
Epoch 956/1000
6/6 - 0s - loss: 0.0305 - mae: 0.0910
Epoch 957/1000
6/6 - 0s - loss: 0.0305 - mae: 0.0908
Epoch 958/1000
6/6 - 1s - loss: 0.0303 - mae: 0.0900
Epoch 959/1000
6/6 - 0s - loss: 0.0303 - mae: 0.0901
Epoch 960/1000
6/6 - 0s - loss: 0.0303 - mae: 0.0904
Epoch 961/1000
6/6 - 1s - loss: 0.0303 - mae: 0.0898
Epoch 962/1000
6/6 - 0s - loss: 0.0302 - mae: 0.0899
Epoch 963/1000
6/6 - 0s - loss: 0.0300 - mae: 0.0894
Epoch 964/1000
6/6 - 1s - loss: 0.0301 - mae: 0.0894
Epoch 965/1000
6/6 - 0s - loss: 0.0304 - mae: 0.0904
Epoch 966/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0897
Epoch 967/1000
6/6 - 0s - loss: 0.0305 - mae: 0.0909
Epoch 968/1000
6/6 - 0s - loss: 0.0303 - mae: 0.0900
Epoch 969/1000
6/6 - 0s - loss: 0.0298 - mae: 0.0896
Epoch 970/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0904
Epoch 971/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0899
Epoch 972/1000
6/6 - 1s - loss: 0.0301 - mae: 0.0906
Epoch 973/1000
6/6 - 0s - loss: 0.0297 - mae: 0.0896
Epoch 974/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0897
Epoch 975/1000
6/6 - 1s - loss: 0.0298 - mae: 0.0900
Epoch 976/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0898
Epoch 977/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0898
Epoch 978/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0910
Epoch 979/1000
6/6 - 0s - loss: 0.0296 - mae: 0.0912
Epoch 980/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0895
Epoch 981/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0890
Epoch 982/1000
6/6 - 0s - loss: 0.0296 - mae: 0.0898
Epoch 983/1000
6/6 - 0s - loss: 0.0296 - mae: 0.0901
Epoch 984/1000
6/6 - 0s - loss: 0.0296 - mae: 0.0901
Epoch 985/1000
6/6 - 0s - loss: 0.0294 - mae: 0.0899
Epoch 986/1000
6/6 - 0s - loss: 0.0296 - mae: 0.0894
Epoch 987/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0891
Epoch 988/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0897
Epoch 989/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0892
Epoch 990/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0898
Epoch 991/1000
6/6 - 0s - loss: 0.0298 - mae: 0.0895
Epoch 992/1000
6/6 - 1s - loss: 0.0297 - mae: 0.0902
Epoch 993/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0902
Epoch 994/1000
6/6 - 1s - loss: 0.0296 - mae: 0.0898
Epoch 995/1000
6/6 - 0s - loss: 0.0296 - mae: 0.0896
Epoch 996/1000
6/6 - 1s - loss: 0.0292 - mae: 0.0898
Epoch 997/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0907
Epoch 998/1000
6/6 - 0s - loss: 0.0292 - mae: 0.0889
Epoch 999/1000
6/6 - 1s - loss: 0.0294 - mae: 0.0901
Epoch 1000/1000
6/6 - 0s - loss: 0.0292 - mae: 0.0902
WARNING:tensorflow:Layer lstm_92 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
WARNING:tensorflow:Layer lstm_93 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/1000
6/6 - 6s - loss: 0.9489 - mae: 0.6544
Epoch 2/1000
6/6 - 0s - loss: 0.9468 - mae: 0.6515
Epoch 3/1000
6/6 - 0s - loss: 0.9432 - mae: 0.6493
Epoch 4/1000
6/6 - 0s - loss: 0.9365 - mae: 0.6462
Epoch 5/1000
6/6 - 0s - loss: 0.9236 - mae: 0.6376
Epoch 6/1000
6/6 - 1s - loss: 0.9009 - mae: 0.6273
Epoch 7/1000
6/6 - 0s - loss: 0.8653 - mae: 0.6123
Epoch 8/1000
6/6 - 0s - loss: 0.8191 - mae: 0.5868
Epoch 9/1000
6/6 - 1s - loss: 0.7572 - mae: 0.5526
Epoch 10/1000
6/6 - 0s - loss: 0.7051 - mae: 0.5244
Epoch 11/1000
6/6 - 0s - loss: 0.6516 - mae: 0.4918
Epoch 12/1000
6/6 - 0s - loss: 0.6149 - mae: 0.4742
Epoch 13/1000
6/6 - 1s - loss: 0.5782 - mae: 0.4629
Epoch 14/1000
6/6 - 0s - loss: 0.5480 - mae: 0.4479
Epoch 15/1000
6/6 - 0s - loss: 0.5267 - mae: 0.4365
Epoch 16/1000
6/6 - 0s - loss: 0.5138 - mae: 0.4263
Epoch 17/1000
6/6 - 0s - loss: 0.5091 - mae: 0.4195
Epoch 18/1000
6/6 - 0s - loss: 0.4929 - mae: 0.4095
Epoch 19/1000
6/6 - 0s - loss: 0.4868 - mae: 0.4034
Epoch 20/1000
6/6 - 1s - loss: 0.4775 - mae: 0.3990
Epoch 21/1000
6/6 - 0s - loss: 0.4763 - mae: 0.3896
Epoch 22/1000
6/6 - 0s - loss: 0.4676 - mae: 0.3935
Epoch 23/1000
6/6 - 0s - loss: 0.4560 - mae: 0.3789
Epoch 24/1000
6/6 - 0s - loss: 0.4497 - mae: 0.3791
Epoch 25/1000
6/6 - 0s - loss: 0.4404 - mae: 0.3735
Epoch 26/1000
6/6 - 0s - loss: 0.4399 - mae: 0.3727
Epoch 27/1000
6/6 - 0s - loss: 0.4297 - mae: 0.3677
Epoch 28/1000
6/6 - 0s - loss: 0.4301 - mae: 0.3680
Epoch 29/1000
6/6 - 0s - loss: 0.4114 - mae: 0.3571
Epoch 30/1000
6/6 - 0s - loss: 0.4104 - mae: 0.3603
Epoch 31/1000
6/6 - 0s - loss: 0.3940 - mae: 0.3492
Epoch 32/1000
6/6 - 0s - loss: 0.3872 - mae: 0.3518
Epoch 33/1000
6/6 - 0s - loss: 0.3721 - mae: 0.3422
Epoch 34/1000
6/6 - 0s - loss: 0.3702 - mae: 0.3400
Epoch 35/1000
6/6 - 0s - loss: 0.3518 - mae: 0.3344
Epoch 36/1000
6/6 - 0s - loss: 0.3376 - mae: 0.3287
Epoch 37/1000
6/6 - 0s - loss: 0.3316 - mae: 0.3278
Epoch 38/1000
6/6 - 0s - loss: 0.3232 - mae: 0.3190
Epoch 39/1000
6/6 - 0s - loss: 0.3089 - mae: 0.3178
Epoch 40/1000
6/6 - 0s - loss: 0.2948 - mae: 0.3112
Epoch 41/1000
6/6 - 0s - loss: 0.2889 - mae: 0.3090
Epoch 42/1000
6/6 - 0s - loss: 0.2796 - mae: 0.3063
Epoch 43/1000
6/6 - 0s - loss: 0.2739 - mae: 0.3016
Epoch 44/1000
6/6 - 1s - loss: 0.2610 - mae: 0.2967
Epoch 45/1000
6/6 - 0s - loss: 0.2639 - mae: 0.2976
Epoch 46/1000
6/6 - 1s - loss: 0.2553 - mae: 0.2962
Epoch 47/1000
6/6 - 1s - loss: 0.2551 - mae: 0.2923
Epoch 48/1000
6/6 - 0s - loss: 0.2465 - mae: 0.2884
Epoch 49/1000
6/6 - 0s - loss: 0.2468 - mae: 0.2890
Epoch 50/1000
6/6 - 0s - loss: 0.2401 - mae: 0.2840
Epoch 51/1000
6/6 - 0s - loss: 0.2338 - mae: 0.2834
Epoch 52/1000
6/6 - 0s - loss: 0.2294 - mae: 0.2778
Epoch 53/1000
6/6 - 0s - loss: 0.2269 - mae: 0.2789
Epoch 54/1000
6/6 - 0s - loss: 0.2227 - mae: 0.2738
Epoch 55/1000
6/6 - 0s - loss: 0.2201 - mae: 0.2751
Epoch 56/1000
6/6 - 0s - loss: 0.2221 - mae: 0.2730
Epoch 57/1000
6/6 - 0s - loss: 0.2192 - mae: 0.2703
Epoch 58/1000
6/6 - 0s - loss: 0.2101 - mae: 0.2698
Epoch 59/1000
6/6 - 0s - loss: 0.2122 - mae: 0.2679
Epoch 60/1000
6/6 - 0s - loss: 0.2109 - mae: 0.2669
Epoch 61/1000
6/6 - 0s - loss: 0.2068 - mae: 0.2644
Epoch 62/1000
6/6 - 0s - loss: 0.2055 - mae: 0.2632
Epoch 63/1000
6/6 - 0s - loss: 0.2006 - mae: 0.2592
Epoch 64/1000
6/6 - 0s - loss: 0.2041 - mae: 0.2604
Epoch 65/1000
6/6 - 0s - loss: 0.1984 - mae: 0.2573
Epoch 66/1000
6/6 - 0s - loss: 0.1930 - mae: 0.2547
Epoch 67/1000
6/6 - 0s - loss: 0.2005 - mae: 0.2571
Epoch 68/1000
6/6 - 0s - loss: 0.1979 - mae: 0.2560
Epoch 69/1000
6/6 - 0s - loss: 0.1957 - mae: 0.2547
Epoch 70/1000
6/6 - 0s - loss: 0.1969 - mae: 0.2556
Epoch 71/1000
6/6 - 0s - loss: 0.1902 - mae: 0.2511
Epoch 72/1000
6/6 - 0s - loss: 0.1841 - mae: 0.2460
Epoch 73/1000
6/6 - 0s - loss: 0.1879 - mae: 0.2486
Epoch 74/1000
6/6 - 0s - loss: 0.1895 - mae: 0.2489
Epoch 75/1000
6/6 - 0s - loss: 0.1892 - mae: 0.2470
Epoch 76/1000
6/6 - 0s - loss: 0.1859 - mae: 0.2441
Epoch 77/1000
6/6 - 0s - loss: 0.1833 - mae: 0.2488
Epoch 78/1000
6/6 - 1s - loss: 0.1809 - mae: 0.2423
Epoch 79/1000
6/6 - 0s - loss: 0.1846 - mae: 0.2457
Epoch 80/1000
6/6 - 0s - loss: 0.1832 - mae: 0.2418
Epoch 81/1000
6/6 - 0s - loss: 0.1795 - mae: 0.2408
Epoch 82/1000
6/6 - 0s - loss: 0.1781 - mae: 0.2422
Epoch 83/1000
6/6 - 0s - loss: 0.1776 - mae: 0.2399
Epoch 84/1000
6/6 - 0s - loss: 0.1776 - mae: 0.2406
Epoch 85/1000
6/6 - 0s - loss: 0.1773 - mae: 0.2395
Epoch 86/1000
6/6 - 0s - loss: 0.1781 - mae: 0.2387
Epoch 87/1000
6/6 - 0s - loss: 0.1762 - mae: 0.2385
Epoch 88/1000
6/6 - 0s - loss: 0.1753 - mae: 0.2371
Epoch 89/1000
6/6 - 0s - loss: 0.1713 - mae: 0.2331
Epoch 90/1000
6/6 - 0s - loss: 0.1725 - mae: 0.2335
Epoch 91/1000
6/6 - 0s - loss: 0.1746 - mae: 0.2353
Epoch 92/1000
6/6 - 1s - loss: 0.1724 - mae: 0.2339
Epoch 93/1000
6/6 - 1s - loss: 0.1706 - mae: 0.2310
Epoch 94/1000
6/6 - 0s - loss: 0.1717 - mae: 0.2333
Epoch 95/1000
6/6 - 0s - loss: 0.1684 - mae: 0.2328
Epoch 96/1000
6/6 - 0s - loss: 0.1687 - mae: 0.2298
Epoch 97/1000
6/6 - 0s - loss: 0.1686 - mae: 0.2295
Epoch 98/1000
6/6 - 0s - loss: 0.1671 - mae: 0.2288
Epoch 99/1000
6/6 - 0s - loss: 0.1668 - mae: 0.2292
Epoch 100/1000
6/6 - 0s - loss: 0.1652 - mae: 0.2279
Epoch 101/1000
6/6 - 0s - loss: 0.1632 - mae: 0.2259
Epoch 102/1000
6/6 - 1s - loss: 0.1698 - mae: 0.2283
Epoch 103/1000
6/6 - 1s - loss: 0.1609 - mae: 0.2246
Epoch 104/1000
6/6 - 0s - loss: 0.1637 - mae: 0.2256
Epoch 105/1000
6/6 - 0s - loss: 0.1656 - mae: 0.2250
Epoch 106/1000
6/6 - 0s - loss: 0.1626 - mae: 0.2248
Epoch 107/1000
6/6 - 0s - loss: 0.1613 - mae: 0.2218
Epoch 108/1000
6/6 - 0s - loss: 0.1594 - mae: 0.2229
Epoch 109/1000
6/6 - 0s - loss: 0.1588 - mae: 0.2207
Epoch 110/1000
6/6 - 0s - loss: 0.1621 - mae: 0.2231
Epoch 111/1000
6/6 - 0s - loss: 0.1611 - mae: 0.2203
Epoch 112/1000
6/6 - 0s - loss: 0.1628 - mae: 0.2226
Epoch 113/1000
6/6 - 0s - loss: 0.1586 - mae: 0.2188
Epoch 114/1000
6/6 - 0s - loss: 0.1609 - mae: 0.2213
Epoch 115/1000
6/6 - 0s - loss: 0.1609 - mae: 0.2189
Epoch 116/1000
6/6 - 0s - loss: 0.1564 - mae: 0.2195
Epoch 117/1000
6/6 - 0s - loss: 0.1565 - mae: 0.2172
Epoch 118/1000
6/6 - 0s - loss: 0.1560 - mae: 0.2179
Epoch 119/1000
6/6 - 0s - loss: 0.1599 - mae: 0.2196
Epoch 120/1000
6/6 - 0s - loss: 0.1566 - mae: 0.2167
Epoch 121/1000
6/6 - 0s - loss: 0.1544 - mae: 0.2167
Epoch 122/1000
6/6 - 0s - loss: 0.1563 - mae: 0.2166
Epoch 123/1000
6/6 - 0s - loss: 0.1541 - mae: 0.2160
Epoch 124/1000
6/6 - 0s - loss: 0.1543 - mae: 0.2148
Epoch 125/1000
6/6 - 0s - loss: 0.1521 - mae: 0.2131
Epoch 126/1000
6/6 - 0s - loss: 0.1543 - mae: 0.2150
Epoch 127/1000
6/6 - 0s - loss: 0.1532 - mae: 0.2139
Epoch 128/1000
6/6 - 0s - loss: 0.1548 - mae: 0.2157
Epoch 129/1000
6/6 - 0s - loss: 0.1533 - mae: 0.2128
Epoch 130/1000
6/6 - 0s - loss: 0.1548 - mae: 0.2149
Epoch 131/1000
6/6 - 0s - loss: 0.1508 - mae: 0.2123
Epoch 132/1000
6/6 - 0s - loss: 0.1522 - mae: 0.2126
Epoch 133/1000
6/6 - 0s - loss: 0.1511 - mae: 0.2122
Epoch 134/1000
6/6 - 0s - loss: 0.1510 - mae: 0.2114
Epoch 135/1000
6/6 - 0s - loss: 0.1518 - mae: 0.2125
Epoch 136/1000
6/6 - 0s - loss: 0.1523 - mae: 0.2125
Epoch 137/1000
6/6 - 0s - loss: 0.1530 - mae: 0.2119
Epoch 138/1000
6/6 - 0s - loss: 0.1527 - mae: 0.2107
Epoch 139/1000
6/6 - 0s - loss: 0.1500 - mae: 0.2101
Epoch 140/1000
6/6 - 0s - loss: 0.1530 - mae: 0.2125
Epoch 141/1000
6/6 - 0s - loss: 0.1471 - mae: 0.2083
Epoch 142/1000
6/6 - 0s - loss: 0.1509 - mae: 0.2101
Epoch 143/1000
6/6 - 0s - loss: 0.1483 - mae: 0.2103
Epoch 144/1000
6/6 - 0s - loss: 0.1495 - mae: 0.2093
Epoch 145/1000
6/6 - 0s - loss: 0.1470 - mae: 0.2063
Epoch 146/1000
6/6 - 0s - loss: 0.1474 - mae: 0.2090
Epoch 147/1000
6/6 - 0s - loss: 0.1483 - mae: 0.2073
Epoch 148/1000
6/6 - 0s - loss: 0.1491 - mae: 0.2072
Epoch 149/1000
6/6 - 0s - loss: 0.1457 - mae: 0.2067
Epoch 150/1000
6/6 - 1s - loss: 0.1482 - mae: 0.2078
Epoch 151/1000
6/6 - 0s - loss: 0.1452 - mae: 0.2073
Epoch 152/1000
6/6 - 0s - loss: 0.1440 - mae: 0.2037
Epoch 153/1000
6/6 - 0s - loss: 0.1462 - mae: 0.2060
Epoch 154/1000
6/6 - 0s - loss: 0.1466 - mae: 0.2063
Epoch 155/1000
6/6 - 0s - loss: 0.1474 - mae: 0.2050
Epoch 156/1000
6/6 - 0s - loss: 0.1415 - mae: 0.2026
Epoch 157/1000
6/6 - 0s - loss: 0.1452 - mae: 0.2047
Epoch 158/1000
6/6 - 0s - loss: 0.1461 - mae: 0.2052
Epoch 159/1000
6/6 - 0s - loss: 0.1463 - mae: 0.2058
Epoch 160/1000
6/6 - 0s - loss: 0.1471 - mae: 0.2048
Epoch 161/1000
6/6 - 0s - loss: 0.1452 - mae: 0.2063
Epoch 162/1000
6/6 - 0s - loss: 0.1444 - mae: 0.2033
Epoch 163/1000
6/6 - 0s - loss: 0.1429 - mae: 0.2031
Epoch 164/1000
6/6 - 0s - loss: 0.1432 - mae: 0.2038
Epoch 165/1000
6/6 - 0s - loss: 0.1435 - mae: 0.2020
Epoch 166/1000
6/6 - 0s - loss: 0.1440 - mae: 0.2027
Epoch 167/1000
6/6 - 0s - loss: 0.1415 - mae: 0.2004
Epoch 168/1000
6/6 - 0s - loss: 0.1421 - mae: 0.2021
Epoch 169/1000
6/6 - 0s - loss: 0.1435 - mae: 0.2015
Epoch 170/1000
6/6 - 0s - loss: 0.1455 - mae: 0.2018
Epoch 171/1000
6/6 - 0s - loss: 0.1426 - mae: 0.2018
Epoch 172/1000
6/6 - 0s - loss: 0.1422 - mae: 0.2012
Epoch 173/1000
6/6 - 0s - loss: 0.1434 - mae: 0.2007
Epoch 174/1000
6/6 - 0s - loss: 0.1438 - mae: 0.2011
Epoch 175/1000
6/6 - 1s - loss: 0.1426 - mae: 0.2012
Epoch 176/1000
6/6 - 0s - loss: 0.1407 - mae: 0.1991
Epoch 177/1000
6/6 - 0s - loss: 0.1433 - mae: 0.2007
Epoch 178/1000
6/6 - 0s - loss: 0.1403 - mae: 0.1995
Epoch 179/1000
6/6 - 0s - loss: 0.1421 - mae: 0.2002
Epoch 180/1000
6/6 - 0s - loss: 0.1403 - mae: 0.1990
Epoch 181/1000
6/6 - 0s - loss: 0.1412 - mae: 0.1997
Epoch 182/1000
6/6 - 0s - loss: 0.1445 - mae: 0.2003
Epoch 183/1000
6/6 - 0s - loss: 0.1403 - mae: 0.1992
Epoch 184/1000
6/6 - 0s - loss: 0.1422 - mae: 0.1983
Epoch 185/1000
6/6 - 0s - loss: 0.1400 - mae: 0.1992
Epoch 186/1000
6/6 - 0s - loss: 0.1401 - mae: 0.1988
Epoch 187/1000
6/6 - 0s - loss: 0.1403 - mae: 0.1981
Epoch 188/1000
6/6 - 0s - loss: 0.1399 - mae: 0.1987
Epoch 189/1000
6/6 - 1s - loss: 0.1389 - mae: 0.1976
Epoch 190/1000
6/6 - 0s - loss: 0.1392 - mae: 0.1973
Epoch 191/1000
6/6 - 0s - loss: 0.1396 - mae: 0.1985
Epoch 192/1000
6/6 - 0s - loss: 0.1417 - mae: 0.1975
Epoch 193/1000
6/6 - 0s - loss: 0.1387 - mae: 0.1986
Epoch 194/1000
6/6 - 0s - loss: 0.1408 - mae: 0.1966
Epoch 195/1000
6/6 - 0s - loss: 0.1379 - mae: 0.1958
Epoch 196/1000
6/6 - 0s - loss: 0.1379 - mae: 0.1977
Epoch 197/1000
6/6 - 0s - loss: 0.1389 - mae: 0.1965
Epoch 198/1000
6/6 - 0s - loss: 0.1392 - mae: 0.1965
Epoch 199/1000
6/6 - 0s - loss: 0.1398 - mae: 0.1967
Epoch 200/1000
6/6 - 0s - loss: 0.1376 - mae: 0.1960
Epoch 201/1000
6/6 - 0s - loss: 0.1380 - mae: 0.1957
Epoch 202/1000
6/6 - 0s - loss: 0.1398 - mae: 0.1974
Epoch 203/1000
6/6 - 0s - loss: 0.1393 - mae: 0.1958
Epoch 204/1000
6/6 - 0s - loss: 0.1374 - mae: 0.1953
Epoch 205/1000
6/6 - 0s - loss: 0.1408 - mae: 0.1972
Epoch 206/1000
6/6 - 0s - loss: 0.1376 - mae: 0.1936
Epoch 207/1000
6/6 - 1s - loss: 0.1340 - mae: 0.1932
Epoch 208/1000
6/6 - 0s - loss: 0.1378 - mae: 0.1944
Epoch 209/1000
6/6 - 0s - loss: 0.1372 - mae: 0.1950
Epoch 210/1000
6/6 - 0s - loss: 0.1386 - mae: 0.1951
Epoch 211/1000
6/6 - 0s - loss: 0.1369 - mae: 0.1949
Epoch 212/1000
6/6 - 0s - loss: 0.1370 - mae: 0.1943
Epoch 213/1000
6/6 - 0s - loss: 0.1360 - mae: 0.1931
Epoch 214/1000
6/6 - 0s - loss: 0.1358 - mae: 0.1927
Epoch 215/1000
6/6 - 0s - loss: 0.1365 - mae: 0.1938
Epoch 216/1000
6/6 - 1s - loss: 0.1377 - mae: 0.1930
Epoch 217/1000
6/6 - 1s - loss: 0.1366 - mae: 0.1946
Epoch 218/1000
6/6 - 0s - loss: 0.1361 - mae: 0.1941
Epoch 219/1000
6/6 - 0s - loss: 0.1349 - mae: 0.1929
Epoch 220/1000
6/6 - 0s - loss: 0.1368 - mae: 0.1941
Epoch 221/1000
6/6 - 1s - loss: 0.1360 - mae: 0.1924
Epoch 222/1000
6/6 - 0s - loss: 0.1367 - mae: 0.1941
Epoch 223/1000
6/6 - 0s - loss: 0.1330 - mae: 0.1904
Epoch 224/1000
6/6 - 0s - loss: 0.1353 - mae: 0.1924
Epoch 225/1000
6/6 - 1s - loss: 0.1345 - mae: 0.1922
Epoch 226/1000
6/6 - 0s - loss: 0.1365 - mae: 0.1937
Epoch 227/1000
6/6 - 0s - loss: 0.1334 - mae: 0.1907
Epoch 228/1000
6/6 - 0s - loss: 0.1330 - mae: 0.1911
Epoch 229/1000
6/6 - 0s - loss: 0.1350 - mae: 0.1927
Epoch 230/1000
6/6 - 0s - loss: 0.1358 - mae: 0.1918
Epoch 231/1000
6/6 - 0s - loss: 0.1349 - mae: 0.1911
Epoch 232/1000
6/6 - 1s - loss: 0.1344 - mae: 0.1912
Epoch 233/1000
6/6 - 0s - loss: 0.1383 - mae: 0.1945
Epoch 234/1000
6/6 - 0s - loss: 0.1345 - mae: 0.1909
Epoch 235/1000
6/6 - 0s - loss: 0.1335 - mae: 0.1913
Epoch 236/1000
6/6 - 0s - loss: 0.1358 - mae: 0.1914
Epoch 237/1000
6/6 - 0s - loss: 0.1373 - mae: 0.1932
Epoch 238/1000
6/6 - 0s - loss: 0.1330 - mae: 0.1910
Epoch 239/1000
6/6 - 1s - loss: 0.1345 - mae: 0.1899
Epoch 240/1000
6/6 - 0s - loss: 0.1337 - mae: 0.1896
Epoch 241/1000
6/6 - 0s - loss: 0.1317 - mae: 0.1898
Epoch 242/1000
6/6 - 0s - loss: 0.1355 - mae: 0.1911
Epoch 243/1000
6/6 - 0s - loss: 0.1332 - mae: 0.1888
Epoch 244/1000
6/6 - 0s - loss: 0.1343 - mae: 0.1903
Epoch 245/1000
6/6 - 0s - loss: 0.1322 - mae: 0.1894
Epoch 246/1000
6/6 - 0s - loss: 0.1349 - mae: 0.1891
Epoch 247/1000
6/6 - 0s - loss: 0.1333 - mae: 0.1904
Epoch 248/1000
6/6 - 0s - loss: 0.1330 - mae: 0.1888
Epoch 249/1000
6/6 - 0s - loss: 0.1310 - mae: 0.1887
Epoch 250/1000
6/6 - 0s - loss: 0.1350 - mae: 0.1903
Epoch 251/1000
6/6 - 0s - loss: 0.1325 - mae: 0.1889
Epoch 252/1000
6/6 - 0s - loss: 0.1326 - mae: 0.1885
Epoch 253/1000
6/6 - 0s - loss: 0.1342 - mae: 0.1900
Epoch 254/1000
6/6 - 0s - loss: 0.1321 - mae: 0.1878
Epoch 255/1000
6/6 - 0s - loss: 0.1359 - mae: 0.1904
Epoch 256/1000
6/6 - 1s - loss: 0.1336 - mae: 0.1889
Epoch 257/1000
6/6 - 1s - loss: 0.1311 - mae: 0.1870
Epoch 258/1000
6/6 - 0s - loss: 0.1324 - mae: 0.1889
Epoch 259/1000
6/6 - 0s - loss: 0.1318 - mae: 0.1874
Epoch 260/1000
6/6 - 0s - loss: 0.1311 - mae: 0.1880
Epoch 261/1000
6/6 - 0s - loss: 0.1313 - mae: 0.1875
Epoch 262/1000
6/6 - 0s - loss: 0.1334 - mae: 0.1881
Epoch 263/1000
6/6 - 0s - loss: 0.1321 - mae: 0.1881
Epoch 264/1000
6/6 - 1s - loss: 0.1311 - mae: 0.1870
Epoch 265/1000
6/6 - 0s - loss: 0.1314 - mae: 0.1871
Epoch 266/1000
6/6 - 0s - loss: 0.1329 - mae: 0.1887
Epoch 267/1000
6/6 - 0s - loss: 0.1315 - mae: 0.1864
Epoch 268/1000
6/6 - 0s - loss: 0.1320 - mae: 0.1866
Epoch 269/1000
6/6 - 0s - loss: 0.1317 - mae: 0.1867
Epoch 270/1000
6/6 - 0s - loss: 0.1334 - mae: 0.1883
Epoch 271/1000
6/6 - 0s - loss: 0.1317 - mae: 0.1873
Epoch 272/1000
6/6 - 0s - loss: 0.1327 - mae: 0.1869
Epoch 273/1000
6/6 - 1s - loss: 0.1315 - mae: 0.1877
Epoch 274/1000
6/6 - 1s - loss: 0.1310 - mae: 0.1859
Epoch 275/1000
6/6 - 1s - loss: 0.1301 - mae: 0.1865
Epoch 276/1000
6/6 - 0s - loss: 0.1299 - mae: 0.1863
Epoch 277/1000
6/6 - 0s - loss: 0.1316 - mae: 0.1870
Epoch 278/1000
6/6 - 1s - loss: 0.1317 - mae: 0.1872
Epoch 279/1000
6/6 - 0s - loss: 0.1300 - mae: 0.1864
Epoch 280/1000
6/6 - 0s - loss: 0.1292 - mae: 0.1864
Epoch 281/1000
6/6 - 0s - loss: 0.1301 - mae: 0.1862
Epoch 282/1000
6/6 - 0s - loss: 0.1316 - mae: 0.1860
Epoch 283/1000
6/6 - 0s - loss: 0.1293 - mae: 0.1851
Epoch 284/1000
6/6 - 0s - loss: 0.1316 - mae: 0.1865
Epoch 285/1000
6/6 - 0s - loss: 0.1317 - mae: 0.1866
Epoch 286/1000
6/6 - 0s - loss: 0.1311 - mae: 0.1860
Epoch 287/1000
6/6 - 0s - loss: 0.1313 - mae: 0.1866
Epoch 288/1000
6/6 - 0s - loss: 0.1308 - mae: 0.1862
Epoch 289/1000
6/6 - 0s - loss: 0.1315 - mae: 0.1853
Epoch 290/1000
6/6 - 0s - loss: 0.1292 - mae: 0.1864
Epoch 291/1000
6/6 - 0s - loss: 0.1292 - mae: 0.1858
Epoch 292/1000
6/6 - 0s - loss: 0.1309 - mae: 0.1847
Epoch 293/1000
6/6 - 0s - loss: 0.1310 - mae: 0.1858
Epoch 294/1000
6/6 - 0s - loss: 0.1282 - mae: 0.1838
Epoch 295/1000
6/6 - 0s - loss: 0.1291 - mae: 0.1852
Epoch 296/1000
6/6 - 0s - loss: 0.1309 - mae: 0.1847
Epoch 297/1000
6/6 - 0s - loss: 0.1309 - mae: 0.1861
Epoch 298/1000
6/6 - 0s - loss: 0.1306 - mae: 0.1854
Epoch 299/1000
6/6 - 0s - loss: 0.1296 - mae: 0.1843
Epoch 300/1000
6/6 - 0s - loss: 0.1288 - mae: 0.1848
Epoch 301/1000
6/6 - 0s - loss: 0.1287 - mae: 0.1846
Epoch 302/1000
6/6 - 0s - loss: 0.1296 - mae: 0.1840
Epoch 303/1000
6/6 - 0s - loss: 0.1296 - mae: 0.1859
Epoch 304/1000
6/6 - 0s - loss: 0.1334 - mae: 0.1854
Epoch 305/1000
6/6 - 0s - loss: 0.1297 - mae: 0.1852
Epoch 306/1000
6/6 - 0s - loss: 0.1290 - mae: 0.1837
Epoch 307/1000
6/6 - 0s - loss: 0.1281 - mae: 0.1846
Epoch 308/1000
6/6 - 0s - loss: 0.1274 - mae: 0.1830
Epoch 309/1000
6/6 - 0s - loss: 0.1303 - mae: 0.1856
Epoch 310/1000
6/6 - 1s - loss: 0.1300 - mae: 0.1842
Epoch 311/1000
6/6 - 0s - loss: 0.1299 - mae: 0.1843
Epoch 312/1000
6/6 - 0s - loss: 0.1290 - mae: 0.1838
Epoch 313/1000
6/6 - 0s - loss: 0.1285 - mae: 0.1841
Epoch 314/1000
6/6 - 0s - loss: 0.1284 - mae: 0.1829
Epoch 315/1000
6/6 - 0s - loss: 0.1295 - mae: 0.1851
Epoch 316/1000
6/6 - 0s - loss: 0.1296 - mae: 0.1843
Epoch 317/1000
6/6 - 1s - loss: 0.1280 - mae: 0.1846
Epoch 318/1000
6/6 - 0s - loss: 0.1289 - mae: 0.1835
Epoch 319/1000
6/6 - 0s - loss: 0.1287 - mae: 0.1839
Epoch 320/1000
6/6 - 1s - loss: 0.1308 - mae: 0.1839
Epoch 321/1000
6/6 - 0s - loss: 0.1287 - mae: 0.1835
Epoch 322/1000
6/6 - 0s - loss: 0.1290 - mae: 0.1852
Epoch 323/1000
6/6 - 0s - loss: 0.1297 - mae: 0.1827
Epoch 324/1000
6/6 - 0s - loss: 0.1290 - mae: 0.1833
Epoch 325/1000
6/6 - 0s - loss: 0.1294 - mae: 0.1847
Epoch 326/1000
6/6 - 0s - loss: 0.1288 - mae: 0.1834
Epoch 327/1000
6/6 - 0s - loss: 0.1273 - mae: 0.1819
Epoch 328/1000
6/6 - 0s - loss: 0.1296 - mae: 0.1829
Epoch 329/1000
6/6 - 0s - loss: 0.1278 - mae: 0.1836
Epoch 330/1000
6/6 - 1s - loss: 0.1299 - mae: 0.1842
Epoch 331/1000
6/6 - 0s - loss: 0.1283 - mae: 0.1823
Epoch 332/1000
6/6 - 0s - loss: 0.1275 - mae: 0.1827
Epoch 333/1000
6/6 - 0s - loss: 0.1268 - mae: 0.1810
Epoch 334/1000
6/6 - 0s - loss: 0.1253 - mae: 0.1812
Epoch 335/1000
6/6 - 0s - loss: 0.1298 - mae: 0.1834
Epoch 336/1000
6/6 - 0s - loss: 0.1283 - mae: 0.1818
Epoch 337/1000
6/6 - 0s - loss: 0.1271 - mae: 0.1831
Epoch 338/1000
6/6 - 0s - loss: 0.1293 - mae: 0.1821
Epoch 339/1000
6/6 - 0s - loss: 0.1302 - mae: 0.1838
Epoch 340/1000
6/6 - 0s - loss: 0.1288 - mae: 0.1828
Epoch 341/1000
6/6 - 0s - loss: 0.1282 - mae: 0.1819
Epoch 342/1000
6/6 - 0s - loss: 0.1269 - mae: 0.1818
Epoch 343/1000
6/6 - 0s - loss: 0.1286 - mae: 0.1821
Epoch 344/1000
6/6 - 0s - loss: 0.1283 - mae: 0.1833
Epoch 345/1000
6/6 - 0s - loss: 0.1280 - mae: 0.1804
Epoch 346/1000
6/6 - 0s - loss: 0.1293 - mae: 0.1823
Epoch 347/1000
6/6 - 0s - loss: 0.1270 - mae: 0.1826
Epoch 348/1000
6/6 - 0s - loss: 0.1273 - mae: 0.1815
Epoch 349/1000
6/6 - 0s - loss: 0.1283 - mae: 0.1828
Epoch 350/1000
6/6 - 0s - loss: 0.1277 - mae: 0.1811
Epoch 351/1000
6/6 - 0s - loss: 0.1271 - mae: 0.1821
Epoch 352/1000
6/6 - 0s - loss: 0.1306 - mae: 0.1826
Epoch 353/1000
6/6 - 0s - loss: 0.1272 - mae: 0.1829
Epoch 354/1000
6/6 - 1s - loss: 0.1265 - mae: 0.1823
Epoch 355/1000
6/6 - 0s - loss: 0.1287 - mae: 0.1813
Epoch 356/1000
6/6 - 0s - loss: 0.1285 - mae: 0.1827
Epoch 357/1000
6/6 - 0s - loss: 0.1273 - mae: 0.1804
Epoch 358/1000
6/6 - 0s - loss: 0.1276 - mae: 0.1816
Epoch 359/1000
6/6 - 0s - loss: 0.1279 - mae: 0.1825
Epoch 360/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1814
Epoch 361/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1810
Epoch 362/1000
6/6 - 0s - loss: 0.1278 - mae: 0.1817
Epoch 363/1000
6/6 - 0s - loss: 0.1270 - mae: 0.1802
Epoch 364/1000
6/6 - 0s - loss: 0.1266 - mae: 0.1823
Epoch 365/1000
6/6 - 0s - loss: 0.1262 - mae: 0.1803
Epoch 366/1000
6/6 - 0s - loss: 0.1257 - mae: 0.1812
Epoch 367/1000
6/6 - 0s - loss: 0.1254 - mae: 0.1798
Epoch 368/1000
6/6 - 0s - loss: 0.1261 - mae: 0.1800
Epoch 369/1000
6/6 - 0s - loss: 0.1274 - mae: 0.1809
Epoch 370/1000
6/6 - 0s - loss: 0.1247 - mae: 0.1785
Epoch 371/1000
6/6 - 0s - loss: 0.1260 - mae: 0.1815
Epoch 372/1000
6/6 - 1s - loss: 0.1260 - mae: 0.1797
Epoch 373/1000
6/6 - 0s - loss: 0.1290 - mae: 0.1816
Epoch 374/1000
6/6 - 0s - loss: 0.1261 - mae: 0.1797
Epoch 375/1000
6/6 - 0s - loss: 0.1275 - mae: 0.1814
Epoch 376/1000
6/6 - 0s - loss: 0.1254 - mae: 0.1793
Epoch 377/1000
6/6 - 0s - loss: 0.1254 - mae: 0.1809
Epoch 378/1000
6/6 - 0s - loss: 0.1264 - mae: 0.1800
Epoch 379/1000
6/6 - 0s - loss: 0.1285 - mae: 0.1813
Epoch 380/1000
6/6 - 0s - loss: 0.1270 - mae: 0.1814
Epoch 381/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1802
Epoch 382/1000
6/6 - 0s - loss: 0.1260 - mae: 0.1805
Epoch 383/1000
6/6 - 0s - loss: 0.1261 - mae: 0.1804
Epoch 384/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1790
Epoch 385/1000
6/6 - 0s - loss: 0.1273 - mae: 0.1803
Epoch 386/1000
6/6 - 1s - loss: 0.1267 - mae: 0.1814
Epoch 387/1000
6/6 - 1s - loss: 0.1264 - mae: 0.1794
Epoch 388/1000
6/6 - 0s - loss: 0.1264 - mae: 0.1798
Epoch 389/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1794
Epoch 390/1000
6/6 - 0s - loss: 0.1251 - mae: 0.1786
Epoch 391/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1791
Epoch 392/1000
6/6 - 0s - loss: 0.1244 - mae: 0.1787
Epoch 393/1000
6/6 - 0s - loss: 0.1237 - mae: 0.1796
Epoch 394/1000
6/6 - 0s - loss: 0.1241 - mae: 0.1794
Epoch 395/1000
6/6 - 1s - loss: 0.1263 - mae: 0.1802
Epoch 396/1000
6/6 - 0s - loss: 0.1269 - mae: 0.1798
Epoch 397/1000
6/6 - 0s - loss: 0.1234 - mae: 0.1784
Epoch 398/1000
6/6 - 1s - loss: 0.1252 - mae: 0.1791
Epoch 399/1000
6/6 - 0s - loss: 0.1261 - mae: 0.1798
Epoch 400/1000
6/6 - 0s - loss: 0.1254 - mae: 0.1792
Epoch 401/1000
6/6 - 0s - loss: 0.1257 - mae: 0.1799
Epoch 402/1000
6/6 - 0s - loss: 0.1242 - mae: 0.1787
Epoch 403/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1790
Epoch 404/1000
6/6 - 0s - loss: 0.1240 - mae: 0.1792
Epoch 405/1000
6/6 - 0s - loss: 0.1256 - mae: 0.1785
Epoch 406/1000
6/6 - 0s - loss: 0.1239 - mae: 0.1788
Epoch 407/1000
6/6 - 1s - loss: 0.1245 - mae: 0.1784
Epoch 408/1000
6/6 - 0s - loss: 0.1248 - mae: 0.1776
Epoch 409/1000
6/6 - 1s - loss: 0.1259 - mae: 0.1798
Epoch 410/1000
6/6 - 0s - loss: 0.1254 - mae: 0.1782
Epoch 411/1000
6/6 - 0s - loss: 0.1263 - mae: 0.1801
Epoch 412/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1787
Epoch 413/1000
6/6 - 0s - loss: 0.1236 - mae: 0.1776
Epoch 414/1000
6/6 - 0s - loss: 0.1238 - mae: 0.1790
Epoch 415/1000
6/6 - 1s - loss: 0.1242 - mae: 0.1777
Epoch 416/1000
6/6 - 1s - loss: 0.1256 - mae: 0.1788
Epoch 417/1000
6/6 - 0s - loss: 0.1251 - mae: 0.1791
Epoch 418/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1778
Epoch 419/1000
6/6 - 0s - loss: 0.1279 - mae: 0.1781
Epoch 420/1000
6/6 - 0s - loss: 0.1238 - mae: 0.1785
Epoch 421/1000
6/6 - 0s - loss: 0.1244 - mae: 0.1780
Epoch 422/1000
6/6 - 0s - loss: 0.1275 - mae: 0.1787
Epoch 423/1000
6/6 - 0s - loss: 0.1246 - mae: 0.1788
Epoch 424/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1776
Epoch 425/1000
6/6 - 0s - loss: 0.1262 - mae: 0.1786
Epoch 426/1000
6/6 - 0s - loss: 0.1231 - mae: 0.1771
Epoch 427/1000
6/6 - 0s - loss: 0.1251 - mae: 0.1783
Epoch 428/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1790
Epoch 429/1000
6/6 - 0s - loss: 0.1241 - mae: 0.1772
Epoch 430/1000
6/6 - 0s - loss: 0.1242 - mae: 0.1787
Epoch 431/1000
6/6 - 0s - loss: 0.1262 - mae: 0.1774
Epoch 432/1000
6/6 - 0s - loss: 0.1238 - mae: 0.1790
Epoch 433/1000
6/6 - 0s - loss: 0.1230 - mae: 0.1769
Epoch 434/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1786
Epoch 435/1000
6/6 - 0s - loss: 0.1236 - mae: 0.1772
Epoch 436/1000
6/6 - 0s - loss: 0.1236 - mae: 0.1763
Epoch 437/1000
6/6 - 0s - loss: 0.1245 - mae: 0.1786
Epoch 438/1000
6/6 - 1s - loss: 0.1243 - mae: 0.1763
Epoch 439/1000
6/6 - 0s - loss: 0.1252 - mae: 0.1784
Epoch 440/1000
6/6 - 0s - loss: 0.1244 - mae: 0.1770
Epoch 441/1000
6/6 - 0s - loss: 0.1246 - mae: 0.1779
Epoch 442/1000
6/6 - 1s - loss: 0.1231 - mae: 0.1767
Epoch 443/1000
6/6 - 0s - loss: 0.1242 - mae: 0.1777
Epoch 444/1000
6/6 - 1s - loss: 0.1227 - mae: 0.1757
Epoch 445/1000
6/6 - 0s - loss: 0.1234 - mae: 0.1772
Epoch 446/1000
6/6 - 0s - loss: 0.1253 - mae: 0.1775
Epoch 447/1000
6/6 - 0s - loss: 0.1237 - mae: 0.1769
Epoch 448/1000
6/6 - 0s - loss: 0.1241 - mae: 0.1771
Epoch 449/1000
6/6 - 0s - loss: 0.1236 - mae: 0.1782
Epoch 450/1000
6/6 - 1s - loss: 0.1234 - mae: 0.1767
Epoch 451/1000
6/6 - 0s - loss: 0.1259 - mae: 0.1783
Epoch 452/1000
6/6 - 0s - loss: 0.1231 - mae: 0.1759
Epoch 453/1000
6/6 - 0s - loss: 0.1230 - mae: 0.1759
Epoch 454/1000
6/6 - 1s - loss: 0.1244 - mae: 0.1764
Epoch 455/1000
6/6 - 0s - loss: 0.1227 - mae: 0.1756
Epoch 456/1000
6/6 - 0s - loss: 0.1234 - mae: 0.1757
Epoch 457/1000
6/6 - 1s - loss: 0.1246 - mae: 0.1767
Epoch 458/1000
6/6 - 0s - loss: 0.1219 - mae: 0.1751
Epoch 459/1000
6/6 - 0s - loss: 0.1255 - mae: 0.1774
Epoch 460/1000
6/6 - 0s - loss: 0.1232 - mae: 0.1762
Epoch 461/1000
6/6 - 0s - loss: 0.1216 - mae: 0.1762
Epoch 462/1000
6/6 - 0s - loss: 0.1231 - mae: 0.1767
Epoch 463/1000
6/6 - 0s - loss: 0.1241 - mae: 0.1755
Epoch 464/1000
6/6 - 0s - loss: 0.1224 - mae: 0.1759
Epoch 465/1000
6/6 - 0s - loss: 0.1240 - mae: 0.1763
Epoch 466/1000
6/6 - 0s - loss: 0.1234 - mae: 0.1754
Epoch 467/1000
6/6 - 0s - loss: 0.1240 - mae: 0.1760
Epoch 468/1000
6/6 - 0s - loss: 0.1248 - mae: 0.1776
Epoch 469/1000
6/6 - 1s - loss: 0.1226 - mae: 0.1750
Epoch 470/1000
6/6 - 1s - loss: 0.1230 - mae: 0.1773
Epoch 471/1000
6/6 - 0s - loss: 0.1225 - mae: 0.1751
Epoch 472/1000
6/6 - 0s - loss: 0.1242 - mae: 0.1760
Epoch 473/1000
6/6 - 0s - loss: 0.1227 - mae: 0.1758
Epoch 474/1000
6/6 - 0s - loss: 0.1234 - mae: 0.1749
Epoch 475/1000
6/6 - 0s - loss: 0.1224 - mae: 0.1761
Epoch 476/1000
6/6 - 0s - loss: 0.1215 - mae: 0.1751
Epoch 477/1000
6/6 - 0s - loss: 0.1245 - mae: 0.1773
Epoch 478/1000
6/6 - 0s - loss: 0.1237 - mae: 0.1759
Epoch 479/1000
6/6 - 0s - loss: 0.1236 - mae: 0.1752
Epoch 480/1000
6/6 - 0s - loss: 0.1228 - mae: 0.1750
Epoch 481/1000
6/6 - 0s - loss: 0.1222 - mae: 0.1760
Epoch 482/1000
6/6 - 0s - loss: 0.1233 - mae: 0.1756
Epoch 483/1000
6/6 - 0s - loss: 0.1213 - mae: 0.1755
Epoch 484/1000
6/6 - 0s - loss: 0.1242 - mae: 0.1762
Epoch 485/1000
6/6 - 0s - loss: 0.1230 - mae: 0.1756
Epoch 486/1000
6/6 - 0s - loss: 0.1226 - mae: 0.1753
Epoch 487/1000
6/6 - 0s - loss: 0.1217 - mae: 0.1752
Epoch 488/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1748
Epoch 489/1000
6/6 - 0s - loss: 0.1247 - mae: 0.1770
Epoch 490/1000
6/6 - 0s - loss: 0.1220 - mae: 0.1745
Epoch 491/1000
6/6 - 0s - loss: 0.1220 - mae: 0.1761
Epoch 492/1000
6/6 - 0s - loss: 0.1244 - mae: 0.1751
Epoch 493/1000
6/6 - 0s - loss: 0.1207 - mae: 0.1745
Epoch 494/1000
6/6 - 0s - loss: 0.1220 - mae: 0.1749
Epoch 495/1000
6/6 - 0s - loss: 0.1234 - mae: 0.1747
Epoch 496/1000
6/6 - 0s - loss: 0.1224 - mae: 0.1766
Epoch 497/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1728
Epoch 498/1000
6/6 - 0s - loss: 0.1221 - mae: 0.1746
Epoch 499/1000
6/6 - 0s - loss: 0.1229 - mae: 0.1756
Epoch 500/1000
6/6 - 0s - loss: 0.1218 - mae: 0.1742
Epoch 501/1000
6/6 - 1s - loss: 0.1218 - mae: 0.1753
Epoch 502/1000
6/6 - 0s - loss: 0.1229 - mae: 0.1740
Epoch 503/1000
6/6 - 0s - loss: 0.1220 - mae: 0.1748
Epoch 504/1000
6/6 - 0s - loss: 0.1223 - mae: 0.1756
Epoch 505/1000
6/6 - 0s - loss: 0.1222 - mae: 0.1746
Epoch 506/1000
6/6 - 0s - loss: 0.1218 - mae: 0.1736
Epoch 507/1000
6/6 - 0s - loss: 0.1217 - mae: 0.1752
Epoch 508/1000
6/6 - 0s - loss: 0.1227 - mae: 0.1735
Epoch 509/1000
6/6 - 0s - loss: 0.1219 - mae: 0.1742
Epoch 510/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1754
Epoch 511/1000
6/6 - 0s - loss: 0.1213 - mae: 0.1730
Epoch 512/1000
6/6 - 1s - loss: 0.1211 - mae: 0.1749
Epoch 513/1000
6/6 - 1s - loss: 0.1233 - mae: 0.1739
Epoch 514/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1745
Epoch 515/1000
6/6 - 0s - loss: 0.1221 - mae: 0.1740
Epoch 516/1000
6/6 - 0s - loss: 0.1225 - mae: 0.1740
Epoch 517/1000
6/6 - 0s - loss: 0.1210 - mae: 0.1733
Epoch 518/1000
6/6 - 0s - loss: 0.1236 - mae: 0.1741
Epoch 519/1000
6/6 - 0s - loss: 0.1222 - mae: 0.1749
Epoch 520/1000
6/6 - 0s - loss: 0.1208 - mae: 0.1743
Epoch 521/1000
6/6 - 0s - loss: 0.1218 - mae: 0.1745
Epoch 522/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1729
Epoch 523/1000
6/6 - 0s - loss: 0.1202 - mae: 0.1736
Epoch 524/1000
6/6 - 0s - loss: 0.1225 - mae: 0.1746
Epoch 525/1000
6/6 - 0s - loss: 0.1223 - mae: 0.1738
Epoch 526/1000
6/6 - 0s - loss: 0.1216 - mae: 0.1737
Epoch 527/1000
6/6 - 0s - loss: 0.1205 - mae: 0.1744
Epoch 528/1000
6/6 - 0s - loss: 0.1230 - mae: 0.1744
Epoch 529/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1728
Epoch 530/1000
6/6 - 0s - loss: 0.1227 - mae: 0.1737
Epoch 531/1000
6/6 - 0s - loss: 0.1216 - mae: 0.1740
Epoch 532/1000
6/6 - 0s - loss: 0.1228 - mae: 0.1737
Epoch 533/1000
6/6 - 1s - loss: 0.1231 - mae: 0.1756
Epoch 534/1000
6/6 - 0s - loss: 0.1232 - mae: 0.1743
Epoch 535/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1740
Epoch 536/1000
6/6 - 0s - loss: 0.1207 - mae: 0.1742
Epoch 537/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1720
Epoch 538/1000
6/6 - 0s - loss: 0.1208 - mae: 0.1727
Epoch 539/1000
6/6 - 0s - loss: 0.1207 - mae: 0.1733
Epoch 540/1000
6/6 - 1s - loss: 0.1215 - mae: 0.1739
Epoch 541/1000
6/6 - 0s - loss: 0.1220 - mae: 0.1742
Epoch 542/1000
6/6 - 0s - loss: 0.1215 - mae: 0.1750
Epoch 543/1000
6/6 - 0s - loss: 0.1237 - mae: 0.1734
Epoch 544/1000
6/6 - 0s - loss: 0.1216 - mae: 0.1730
Epoch 545/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1734
Epoch 546/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1727
Epoch 547/1000
6/6 - 0s - loss: 0.1210 - mae: 0.1733
Epoch 548/1000
6/6 - 0s - loss: 0.1213 - mae: 0.1734
Epoch 549/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1745
Epoch 550/1000
6/6 - 1s - loss: 0.1210 - mae: 0.1731
Epoch 551/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1740
Epoch 552/1000
6/6 - 0s - loss: 0.1193 - mae: 0.1709
Epoch 553/1000
6/6 - 1s - loss: 0.1207 - mae: 0.1733
Epoch 554/1000
6/6 - 1s - loss: 0.1209 - mae: 0.1725
Epoch 555/1000
6/6 - 0s - loss: 0.1221 - mae: 0.1736
Epoch 556/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1716
Epoch 557/1000
6/6 - 0s - loss: 0.1240 - mae: 0.1739
Epoch 558/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1724
Epoch 559/1000
6/6 - 1s - loss: 0.1219 - mae: 0.1733
Epoch 560/1000
6/6 - 1s - loss: 0.1207 - mae: 0.1727
Epoch 561/1000
6/6 - 0s - loss: 0.1215 - mae: 0.1733
Epoch 562/1000
6/6 - 0s - loss: 0.1188 - mae: 0.1722
Epoch 563/1000
6/6 - 0s - loss: 0.1218 - mae: 0.1717
Epoch 564/1000
6/6 - 0s - loss: 0.1215 - mae: 0.1730
Epoch 565/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1727
Epoch 566/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1729
Epoch 567/1000
6/6 - 0s - loss: 0.1210 - mae: 0.1727
Epoch 568/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1728
Epoch 569/1000
6/6 - 0s - loss: 0.1216 - mae: 0.1720
Epoch 570/1000
6/6 - 0s - loss: 0.1199 - mae: 0.1721
Epoch 571/1000
6/6 - 0s - loss: 0.1209 - mae: 0.1721
Epoch 572/1000
6/6 - 0s - loss: 0.1222 - mae: 0.1731
Epoch 573/1000
6/6 - 0s - loss: 0.1207 - mae: 0.1728
Epoch 574/1000
6/6 - 0s - loss: 0.1209 - mae: 0.1714
Epoch 575/1000
6/6 - 1s - loss: 0.1201 - mae: 0.1731
Epoch 576/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1716
Epoch 577/1000
6/6 - 0s - loss: 0.1217 - mae: 0.1749
Epoch 578/1000
6/6 - 0s - loss: 0.1208 - mae: 0.1727
Epoch 579/1000
6/6 - 1s - loss: 0.1204 - mae: 0.1724
Epoch 580/1000
6/6 - 0s - loss: 0.1199 - mae: 0.1720
Epoch 581/1000
6/6 - 0s - loss: 0.1218 - mae: 0.1729
Epoch 582/1000
6/6 - 0s - loss: 0.1195 - mae: 0.1721
Epoch 583/1000
6/6 - 0s - loss: 0.1211 - mae: 0.1716
Epoch 584/1000
6/6 - 0s - loss: 0.1195 - mae: 0.1716
Epoch 585/1000
6/6 - 0s - loss: 0.1199 - mae: 0.1721
Epoch 586/1000
6/6 - 0s - loss: 0.1213 - mae: 0.1723
Epoch 587/1000
6/6 - 0s - loss: 0.1218 - mae: 0.1725
Epoch 588/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1729
Epoch 589/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1718
Epoch 590/1000
6/6 - 0s - loss: 0.1210 - mae: 0.1724
Epoch 591/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1722
Epoch 592/1000
6/6 - 0s - loss: 0.1214 - mae: 0.1719
Epoch 593/1000
6/6 - 0s - loss: 0.1205 - mae: 0.1737
Epoch 594/1000
6/6 - 0s - loss: 0.1207 - mae: 0.1728
Epoch 595/1000
6/6 - 0s - loss: 0.1230 - mae: 0.1730
Epoch 596/1000
6/6 - 0s - loss: 0.1207 - mae: 0.1711
Epoch 597/1000
6/6 - 0s - loss: 0.1203 - mae: 0.1716
Epoch 598/1000
6/6 - 0s - loss: 0.1180 - mae: 0.1711
Epoch 599/1000
6/6 - 0s - loss: 0.1202 - mae: 0.1718
Epoch 600/1000
6/6 - 0s - loss: 0.1210 - mae: 0.1714
Epoch 601/1000
6/6 - 0s - loss: 0.1198 - mae: 0.1708
Epoch 602/1000
6/6 - 0s - loss: 0.1192 - mae: 0.1718
Epoch 603/1000
6/6 - 0s - loss: 0.1209 - mae: 0.1721
Epoch 604/1000
6/6 - 0s - loss: 0.1186 - mae: 0.1717
Epoch 605/1000
6/6 - 0s - loss: 0.1223 - mae: 0.1723
Epoch 606/1000
6/6 - 0s - loss: 0.1190 - mae: 0.1711
Epoch 607/1000
6/6 - 0s - loss: 0.1209 - mae: 0.1717
Epoch 608/1000
6/6 - 0s - loss: 0.1198 - mae: 0.1708
Epoch 609/1000
6/6 - 0s - loss: 0.1211 - mae: 0.1726
Epoch 610/1000
6/6 - 1s - loss: 0.1210 - mae: 0.1715
Epoch 611/1000
6/6 - 0s - loss: 0.1213 - mae: 0.1717
Epoch 612/1000
6/6 - 0s - loss: 0.1209 - mae: 0.1728
Epoch 613/1000
6/6 - 0s - loss: 0.1201 - mae: 0.1713
Epoch 614/1000
6/6 - 0s - loss: 0.1219 - mae: 0.1725
Epoch 615/1000
6/6 - 0s - loss: 0.1184 - mae: 0.1704
Epoch 616/1000
6/6 - 0s - loss: 0.1196 - mae: 0.1715
Epoch 617/1000
6/6 - 0s - loss: 0.1197 - mae: 0.1713
Epoch 618/1000
6/6 - 0s - loss: 0.1186 - mae: 0.1713
Epoch 619/1000
6/6 - 0s - loss: 0.1202 - mae: 0.1720
Epoch 620/1000
6/6 - 0s - loss: 0.1190 - mae: 0.1712
Epoch 621/1000
6/6 - 0s - loss: 0.1219 - mae: 0.1708
Epoch 622/1000
6/6 - 1s - loss: 0.1199 - mae: 0.1718
Epoch 623/1000
6/6 - 0s - loss: 0.1198 - mae: 0.1702
Epoch 624/1000
6/6 - 0s - loss: 0.1186 - mae: 0.1712
Epoch 625/1000
6/6 - 0s - loss: 0.1182 - mae: 0.1708
Epoch 626/1000
6/6 - 0s - loss: 0.1195 - mae: 0.1710
Epoch 627/1000
6/6 - 0s - loss: 0.1222 - mae: 0.1722
Epoch 628/1000
6/6 - 0s - loss: 0.1219 - mae: 0.1717
Epoch 629/1000
6/6 - 0s - loss: 0.1194 - mae: 0.1717
Epoch 630/1000
6/6 - 0s - loss: 0.1194 - mae: 0.1704
Epoch 631/1000
6/6 - 0s - loss: 0.1201 - mae: 0.1704
Epoch 632/1000
6/6 - 0s - loss: 0.1187 - mae: 0.1706
Epoch 633/1000
6/6 - 0s - loss: 0.1195 - mae: 0.1713
Epoch 634/1000
6/6 - 0s - loss: 0.1199 - mae: 0.1722
Epoch 635/1000
6/6 - 0s - loss: 0.1219 - mae: 0.1717
Epoch 636/1000
6/6 - 0s - loss: 0.1196 - mae: 0.1699
Epoch 637/1000
6/6 - 0s - loss: 0.1206 - mae: 0.1713
Epoch 638/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1721
Epoch 639/1000
6/6 - 0s - loss: 0.1189 - mae: 0.1707
Epoch 640/1000
6/6 - 1s - loss: 0.1187 - mae: 0.1707
Epoch 641/1000
6/6 - 0s - loss: 0.1187 - mae: 0.1703
Epoch 642/1000
6/6 - 0s - loss: 0.1191 - mae: 0.1707
Epoch 643/1000
6/6 - 0s - loss: 0.1188 - mae: 0.1708
Epoch 644/1000
6/6 - 0s - loss: 0.1209 - mae: 0.1714
Epoch 645/1000
6/6 - 0s - loss: 0.1202 - mae: 0.1711
Epoch 646/1000
6/6 - 0s - loss: 0.1199 - mae: 0.1707
Epoch 647/1000
6/6 - 1s - loss: 0.1189 - mae: 0.1711
Epoch 648/1000
6/6 - 0s - loss: 0.1204 - mae: 0.1699
WARNING:tensorflow:Layer lstm_94 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
WARNING:tensorflow:Layer lstm_95 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/1000
6/6 - 6s - loss: 0.9370 - mae: 0.6547
Epoch 2/1000
6/6 - 0s - loss: 0.8878 - mae: 0.6389
Epoch 3/1000
6/6 - 0s - loss: 0.7994 - mae: 0.6123
Epoch 4/1000
6/6 - 0s - loss: 0.7429 - mae: 0.5925
Epoch 5/1000
6/6 - 0s - loss: 0.6997 - mae: 0.5727
Epoch 6/1000
6/6 - 1s - loss: 0.6552 - mae: 0.5599
Epoch 7/1000
6/6 - 0s - loss: 0.6089 - mae: 0.5437
Epoch 8/1000
6/6 - 0s - loss: 0.5626 - mae: 0.5263
Epoch 9/1000
6/6 - 0s - loss: 0.5123 - mae: 0.5027
Epoch 10/1000
6/6 - 0s - loss: 0.4605 - mae: 0.4760
Epoch 11/1000
6/6 - 0s - loss: 0.4121 - mae: 0.4492
Epoch 12/1000
6/6 - 0s - loss: 0.3653 - mae: 0.4203
Epoch 13/1000
6/6 - 0s - loss: 0.3272 - mae: 0.3917
Epoch 14/1000
6/6 - 0s - loss: 0.2984 - mae: 0.3708
Epoch 15/1000
6/6 - 0s - loss: 0.2739 - mae: 0.3529
Epoch 16/1000
6/6 - 0s - loss: 0.2569 - mae: 0.3374
Epoch 17/1000
6/6 - 0s - loss: 0.2409 - mae: 0.3225
Epoch 18/1000
6/6 - 0s - loss: 0.2265 - mae: 0.3085
Epoch 19/1000
6/6 - 0s - loss: 0.2160 - mae: 0.2970
Epoch 20/1000
6/6 - 0s - loss: 0.2050 - mae: 0.2860
Epoch 21/1000
6/6 - 0s - loss: 0.1989 - mae: 0.2767
Epoch 22/1000
6/6 - 0s - loss: 0.1891 - mae: 0.2664
Epoch 23/1000
6/6 - 0s - loss: 0.1821 - mae: 0.2569
Epoch 24/1000
6/6 - 0s - loss: 0.1774 - mae: 0.2508
Epoch 25/1000
6/6 - 0s - loss: 0.1736 - mae: 0.2438
Epoch 26/1000
6/6 - 0s - loss: 0.1691 - mae: 0.2381
Epoch 27/1000
6/6 - 0s - loss: 0.1643 - mae: 0.2320
Epoch 28/1000
6/6 - 1s - loss: 0.1603 - mae: 0.2257
Epoch 29/1000
6/6 - 1s - loss: 0.1584 - mae: 0.2217
Epoch 30/1000
6/6 - 0s - loss: 0.1539 - mae: 0.2156
Epoch 31/1000
6/6 - 0s - loss: 0.1544 - mae: 0.2129
Epoch 32/1000
6/6 - 0s - loss: 0.1486 - mae: 0.2091
Epoch 33/1000
6/6 - 0s - loss: 0.1466 - mae: 0.2044
Epoch 34/1000
6/6 - 0s - loss: 0.1471 - mae: 0.2032
Epoch 35/1000
6/6 - 0s - loss: 0.1429 - mae: 0.2001
Epoch 36/1000
6/6 - 0s - loss: 0.1412 - mae: 0.1959
Epoch 37/1000
6/6 - 0s - loss: 0.1380 - mae: 0.1932
Epoch 38/1000
6/6 - 0s - loss: 0.1377 - mae: 0.1921
Epoch 39/1000
6/6 - 0s - loss: 0.1354 - mae: 0.1880
Epoch 40/1000
6/6 - 0s - loss: 0.1358 - mae: 0.1884
Epoch 41/1000
6/6 - 0s - loss: 0.1320 - mae: 0.1849
Epoch 42/1000
6/6 - 0s - loss: 0.1320 - mae: 0.1857
Epoch 43/1000
6/6 - 0s - loss: 0.1307 - mae: 0.1830
Epoch 44/1000
6/6 - 0s - loss: 0.1292 - mae: 0.1811
Epoch 45/1000
6/6 - 0s - loss: 0.1275 - mae: 0.1795
Epoch 46/1000
6/6 - 0s - loss: 0.1266 - mae: 0.1777
Epoch 47/1000
6/6 - 0s - loss: 0.1258 - mae: 0.1779
Epoch 48/1000
6/6 - 0s - loss: 0.1227 - mae: 0.1764
Epoch 49/1000
6/6 - 0s - loss: 0.1208 - mae: 0.1742
Epoch 50/1000
6/6 - 0s - loss: 0.1212 - mae: 0.1750
Epoch 51/1000
6/6 - 0s - loss: 0.1199 - mae: 0.1733
Epoch 52/1000
6/6 - 0s - loss: 0.1183 - mae: 0.1713
Epoch 53/1000
6/6 - 0s - loss: 0.1184 - mae: 0.1702
Epoch 54/1000
6/6 - 0s - loss: 0.1171 - mae: 0.1699
Epoch 55/1000
6/6 - 0s - loss: 0.1143 - mae: 0.1676
Epoch 56/1000
6/6 - 0s - loss: 0.1141 - mae: 0.1660
Epoch 57/1000
6/6 - 0s - loss: 0.1137 - mae: 0.1661
Epoch 58/1000
6/6 - 0s - loss: 0.1123 - mae: 0.1657
Epoch 59/1000
6/6 - 0s - loss: 0.1120 - mae: 0.1641
Epoch 60/1000
6/6 - 0s - loss: 0.1105 - mae: 0.1637
Epoch 61/1000
6/6 - 0s - loss: 0.1102 - mae: 0.1629
Epoch 62/1000
6/6 - 0s - loss: 0.1086 - mae: 0.1620
Epoch 63/1000
6/6 - 0s - loss: 0.1079 - mae: 0.1607
Epoch 64/1000
6/6 - 0s - loss: 0.1092 - mae: 0.1610
Epoch 65/1000
6/6 - 0s - loss: 0.1061 - mae: 0.1595
Epoch 66/1000
6/6 - 0s - loss: 0.1071 - mae: 0.1614
Epoch 67/1000
6/6 - 0s - loss: 0.1072 - mae: 0.1614
Epoch 68/1000
6/6 - 1s - loss: 0.1067 - mae: 0.1604
Epoch 69/1000
6/6 - 0s - loss: 0.1055 - mae: 0.1584
Epoch 70/1000
6/6 - 0s - loss: 0.1046 - mae: 0.1580
Epoch 71/1000
6/6 - 0s - loss: 0.1045 - mae: 0.1574
Epoch 72/1000
6/6 - 0s - loss: 0.1033 - mae: 0.1569
Epoch 73/1000
6/6 - 0s - loss: 0.1029 - mae: 0.1560
Epoch 74/1000
6/6 - 0s - loss: 0.1035 - mae: 0.1560
Epoch 75/1000
6/6 - 0s - loss: 0.1017 - mae: 0.1547
Epoch 76/1000
6/6 - 0s - loss: 0.1023 - mae: 0.1547
Epoch 77/1000
6/6 - 0s - loss: 0.1009 - mae: 0.1545
Epoch 78/1000
6/6 - 0s - loss: 0.1013 - mae: 0.1541
Epoch 79/1000
6/6 - 1s - loss: 0.0996 - mae: 0.1529
Epoch 80/1000
6/6 - 0s - loss: 0.0988 - mae: 0.1521
Epoch 81/1000
6/6 - 0s - loss: 0.0979 - mae: 0.1514
Epoch 82/1000
6/6 - 0s - loss: 0.0991 - mae: 0.1523
Epoch 83/1000
6/6 - 0s - loss: 0.0972 - mae: 0.1500
Epoch 84/1000
6/6 - 0s - loss: 0.0966 - mae: 0.1498
Epoch 85/1000
6/6 - 0s - loss: 0.0980 - mae: 0.1510
Epoch 86/1000
6/6 - 1s - loss: 0.0956 - mae: 0.1488
Epoch 87/1000
6/6 - 0s - loss: 0.0983 - mae: 0.1500
Epoch 88/1000
6/6 - 0s - loss: 0.0948 - mae: 0.1478
Epoch 89/1000
6/6 - 0s - loss: 0.0948 - mae: 0.1477
Epoch 90/1000
6/6 - 0s - loss: 0.0950 - mae: 0.1474
Epoch 91/1000
6/6 - 0s - loss: 0.0950 - mae: 0.1482
Epoch 92/1000
6/6 - 0s - loss: 0.0963 - mae: 0.1485
Epoch 93/1000
6/6 - 0s - loss: 0.0946 - mae: 0.1466
Epoch 94/1000
6/6 - 0s - loss: 0.0936 - mae: 0.1462
Epoch 95/1000
6/6 - 0s - loss: 0.0934 - mae: 0.1452
Epoch 96/1000
6/6 - 0s - loss: 0.0947 - mae: 0.1460
Epoch 97/1000
6/6 - 0s - loss: 0.0938 - mae: 0.1458
Epoch 98/1000
6/6 - 0s - loss: 0.0935 - mae: 0.1457
Epoch 99/1000
6/6 - 0s - loss: 0.0928 - mae: 0.1458
Epoch 100/1000
6/6 - 0s - loss: 0.0920 - mae: 0.1439
Epoch 101/1000
6/6 - 0s - loss: 0.0914 - mae: 0.1444
Epoch 102/1000
6/6 - 0s - loss: 0.0909 - mae: 0.1427
Epoch 103/1000
6/6 - 0s - loss: 0.0919 - mae: 0.1429
Epoch 104/1000
6/6 - 0s - loss: 0.0915 - mae: 0.1418
Epoch 105/1000
6/6 - 0s - loss: 0.0904 - mae: 0.1417
Epoch 106/1000
6/6 - 0s - loss: 0.0897 - mae: 0.1422
Epoch 107/1000
6/6 - 0s - loss: 0.0903 - mae: 0.1419
Epoch 108/1000
6/6 - 0s - loss: 0.0912 - mae: 0.1426
Epoch 109/1000
6/6 - 1s - loss: 0.0898 - mae: 0.1404
Epoch 110/1000
6/6 - 0s - loss: 0.0894 - mae: 0.1413
Epoch 111/1000
6/6 - 0s - loss: 0.0900 - mae: 0.1408
Epoch 112/1000
6/6 - 0s - loss: 0.0891 - mae: 0.1413
Epoch 113/1000
6/6 - 0s - loss: 0.0881 - mae: 0.1395
Epoch 114/1000
6/6 - 0s - loss: 0.0882 - mae: 0.1393
Epoch 115/1000
6/6 - 0s - loss: 0.0873 - mae: 0.1388
Epoch 116/1000
6/6 - 0s - loss: 0.0882 - mae: 0.1382
Epoch 117/1000
6/6 - 0s - loss: 0.0873 - mae: 0.1382
Epoch 118/1000
6/6 - 0s - loss: 0.0863 - mae: 0.1392
Epoch 119/1000
6/6 - 1s - loss: 0.0871 - mae: 0.1381
Epoch 120/1000
6/6 - 0s - loss: 0.0861 - mae: 0.1375
Epoch 121/1000
6/6 - 1s - loss: 0.0868 - mae: 0.1367
Epoch 122/1000
6/6 - 1s - loss: 0.0874 - mae: 0.1385
Epoch 123/1000
6/6 - 0s - loss: 0.0877 - mae: 0.1388
Epoch 124/1000
6/6 - 0s - loss: 0.0860 - mae: 0.1375
Epoch 125/1000
6/6 - 0s - loss: 0.0850 - mae: 0.1358
Epoch 126/1000
6/6 - 0s - loss: 0.0860 - mae: 0.1369
Epoch 127/1000
6/6 - 0s - loss: 0.0851 - mae: 0.1358
Epoch 128/1000
6/6 - 0s - loss: 0.0862 - mae: 0.1377
Epoch 129/1000
6/6 - 0s - loss: 0.0868 - mae: 0.1396
Epoch 130/1000
6/6 - 0s - loss: 0.0860 - mae: 0.1373
Epoch 131/1000
6/6 - 0s - loss: 0.0850 - mae: 0.1355
Epoch 132/1000
6/6 - 0s - loss: 0.0841 - mae: 0.1350
Epoch 133/1000
6/6 - 0s - loss: 0.0846 - mae: 0.1351
Epoch 134/1000
6/6 - 0s - loss: 0.0848 - mae: 0.1354
Epoch 135/1000
6/6 - 1s - loss: 0.0842 - mae: 0.1338
Epoch 136/1000
6/6 - 0s - loss: 0.0837 - mae: 0.1335
Epoch 137/1000
6/6 - 0s - loss: 0.0850 - mae: 0.1340
Epoch 138/1000
6/6 - 0s - loss: 0.0843 - mae: 0.1339
Epoch 139/1000
6/6 - 0s - loss: 0.0848 - mae: 0.1337
Epoch 140/1000
6/6 - 1s - loss: 0.0846 - mae: 0.1354
Epoch 141/1000
6/6 - 0s - loss: 0.0844 - mae: 0.1337
Epoch 142/1000
6/6 - 1s - loss: 0.0834 - mae: 0.1332
Epoch 143/1000
6/6 - 0s - loss: 0.0835 - mae: 0.1334
Epoch 144/1000
6/6 - 0s - loss: 0.0822 - mae: 0.1326
Epoch 145/1000
6/6 - 0s - loss: 0.0825 - mae: 0.1325
Epoch 146/1000
6/6 - 0s - loss: 0.0826 - mae: 0.1327
Epoch 147/1000
6/6 - 0s - loss: 0.0826 - mae: 0.1331
Epoch 148/1000
6/6 - 0s - loss: 0.0832 - mae: 0.1332
Epoch 149/1000
6/6 - 0s - loss: 0.0811 - mae: 0.1321
Epoch 150/1000
6/6 - 0s - loss: 0.0817 - mae: 0.1314
Epoch 151/1000
6/6 - 0s - loss: 0.0820 - mae: 0.1318
Epoch 152/1000
6/6 - 0s - loss: 0.0814 - mae: 0.1312
Epoch 153/1000
6/6 - 0s - loss: 0.0820 - mae: 0.1321
Epoch 154/1000
6/6 - 0s - loss: 0.0813 - mae: 0.1306
Epoch 155/1000
6/6 - 1s - loss: 0.0810 - mae: 0.1310
Epoch 156/1000
6/6 - 0s - loss: 0.0821 - mae: 0.1312
Epoch 157/1000
6/6 - 0s - loss: 0.0806 - mae: 0.1311
Epoch 158/1000
6/6 - 0s - loss: 0.0817 - mae: 0.1306
Epoch 159/1000
6/6 - 0s - loss: 0.0812 - mae: 0.1308
Epoch 160/1000
6/6 - 0s - loss: 0.0808 - mae: 0.1302
Epoch 161/1000
6/6 - 0s - loss: 0.0805 - mae: 0.1304
Epoch 162/1000
6/6 - 0s - loss: 0.0802 - mae: 0.1304
Epoch 163/1000
6/6 - 0s - loss: 0.0801 - mae: 0.1295
Epoch 164/1000
6/6 - 0s - loss: 0.0807 - mae: 0.1304
Epoch 165/1000
6/6 - 0s - loss: 0.0806 - mae: 0.1290
Epoch 166/1000
6/6 - 0s - loss: 0.0797 - mae: 0.1290
Epoch 167/1000
6/6 - 0s - loss: 0.0802 - mae: 0.1293
Epoch 168/1000
6/6 - 0s - loss: 0.0792 - mae: 0.1290
Epoch 169/1000
6/6 - 0s - loss: 0.0806 - mae: 0.1296
Epoch 170/1000
6/6 - 0s - loss: 0.0791 - mae: 0.1289
Epoch 171/1000
6/6 - 0s - loss: 0.0790 - mae: 0.1285
Epoch 172/1000
6/6 - 0s - loss: 0.0787 - mae: 0.1276
Epoch 173/1000
6/6 - 0s - loss: 0.0802 - mae: 0.1286
Epoch 174/1000
6/6 - 0s - loss: 0.0790 - mae: 0.1283
Epoch 175/1000
6/6 - 0s - loss: 0.0795 - mae: 0.1277
Epoch 176/1000
6/6 - 0s - loss: 0.0790 - mae: 0.1282
Epoch 177/1000
6/6 - 0s - loss: 0.0788 - mae: 0.1273
Epoch 178/1000
6/6 - 0s - loss: 0.0786 - mae: 0.1281
Epoch 179/1000
6/6 - 0s - loss: 0.0793 - mae: 0.1279
Epoch 180/1000
6/6 - 0s - loss: 0.0791 - mae: 0.1280
Epoch 181/1000
6/6 - 0s - loss: 0.0782 - mae: 0.1271
Epoch 182/1000
6/6 - 0s - loss: 0.0782 - mae: 0.1268
Epoch 183/1000
6/6 - 0s - loss: 0.0783 - mae: 0.1277
Epoch 184/1000
6/6 - 0s - loss: 0.0778 - mae: 0.1275
Epoch 185/1000
6/6 - 0s - loss: 0.0774 - mae: 0.1263
Epoch 186/1000
6/6 - 0s - loss: 0.0783 - mae: 0.1274
Epoch 187/1000
6/6 - 0s - loss: 0.0785 - mae: 0.1292
Epoch 188/1000
6/6 - 0s - loss: 0.0777 - mae: 0.1270
Epoch 189/1000
6/6 - 0s - loss: 0.0772 - mae: 0.1265
Epoch 190/1000
6/6 - 0s - loss: 0.0781 - mae: 0.1269
Epoch 191/1000
6/6 - 0s - loss: 0.0776 - mae: 0.1275
Epoch 192/1000
6/6 - 0s - loss: 0.0781 - mae: 0.1291
Epoch 193/1000
6/6 - 0s - loss: 0.0773 - mae: 0.1264
Epoch 194/1000
6/6 - 1s - loss: 0.0772 - mae: 0.1263
Epoch 195/1000
6/6 - 0s - loss: 0.0773 - mae: 0.1266
Epoch 196/1000
6/6 - 0s - loss: 0.0771 - mae: 0.1259
Epoch 197/1000
6/6 - 1s - loss: 0.0761 - mae: 0.1256
Epoch 198/1000
6/6 - 0s - loss: 0.0764 - mae: 0.1248
Epoch 199/1000
6/6 - 1s - loss: 0.0761 - mae: 0.1248
Epoch 200/1000
6/6 - 0s - loss: 0.0755 - mae: 0.1241
Epoch 201/1000
6/6 - 0s - loss: 0.0763 - mae: 0.1251
Epoch 202/1000
6/6 - 0s - loss: 0.0765 - mae: 0.1253
Epoch 203/1000
6/6 - 0s - loss: 0.0764 - mae: 0.1252
Epoch 204/1000
6/6 - 0s - loss: 0.0757 - mae: 0.1241
Epoch 205/1000
6/6 - 0s - loss: 0.0750 - mae: 0.1241
Epoch 206/1000
6/6 - 0s - loss: 0.0753 - mae: 0.1236
Epoch 207/1000
6/6 - 0s - loss: 0.0754 - mae: 0.1247
Epoch 208/1000
6/6 - 0s - loss: 0.0754 - mae: 0.1240
Epoch 209/1000
6/6 - 0s - loss: 0.0760 - mae: 0.1251
Epoch 210/1000
6/6 - 0s - loss: 0.0763 - mae: 0.1260
Epoch 211/1000
6/6 - 0s - loss: 0.0755 - mae: 0.1240
Epoch 212/1000
6/6 - 0s - loss: 0.0751 - mae: 0.1238
Epoch 213/1000
6/6 - 0s - loss: 0.0750 - mae: 0.1247
Epoch 214/1000
6/6 - 0s - loss: 0.0750 - mae: 0.1241
Epoch 215/1000
6/6 - 0s - loss: 0.0741 - mae: 0.1234
Epoch 216/1000
6/6 - 0s - loss: 0.0750 - mae: 0.1235
Epoch 217/1000
6/6 - 1s - loss: 0.0748 - mae: 0.1238
Epoch 218/1000
6/6 - 0s - loss: 0.0756 - mae: 0.1266
Epoch 219/1000
6/6 - 0s - loss: 0.0747 - mae: 0.1244
Epoch 220/1000
6/6 - 0s - loss: 0.0737 - mae: 0.1224
Epoch 221/1000
6/6 - 0s - loss: 0.0740 - mae: 0.1234
Epoch 222/1000
6/6 - 0s - loss: 0.0744 - mae: 0.1225
Epoch 223/1000
6/6 - 0s - loss: 0.0734 - mae: 0.1223
Epoch 224/1000
6/6 - 1s - loss: 0.0742 - mae: 0.1229
Epoch 225/1000
6/6 - 1s - loss: 0.0737 - mae: 0.1224
Epoch 226/1000
6/6 - 0s - loss: 0.0743 - mae: 0.1228
Epoch 227/1000
6/6 - 0s - loss: 0.0735 - mae: 0.1223
Epoch 228/1000
6/6 - 0s - loss: 0.0734 - mae: 0.1224
Epoch 229/1000
6/6 - 0s - loss: 0.0726 - mae: 0.1209
Epoch 230/1000
6/6 - 0s - loss: 0.0734 - mae: 0.1221
Epoch 231/1000
6/6 - 0s - loss: 0.0735 - mae: 0.1224
Epoch 232/1000
6/6 - 0s - loss: 0.0729 - mae: 0.1215
Epoch 233/1000
6/6 - 0s - loss: 0.0734 - mae: 0.1215
Epoch 234/1000
6/6 - 0s - loss: 0.0732 - mae: 0.1216
Epoch 235/1000
6/6 - 0s - loss: 0.0723 - mae: 0.1210
Epoch 236/1000
6/6 - 0s - loss: 0.0725 - mae: 0.1212
Epoch 237/1000
6/6 - 0s - loss: 0.0722 - mae: 0.1219
Epoch 238/1000
6/6 - 0s - loss: 0.0721 - mae: 0.1209
Epoch 239/1000
6/6 - 0s - loss: 0.0727 - mae: 0.1214
Epoch 240/1000
6/6 - 0s - loss: 0.0724 - mae: 0.1211
Epoch 241/1000
6/6 - 0s - loss: 0.0721 - mae: 0.1203
Epoch 242/1000
6/6 - 1s - loss: 0.0721 - mae: 0.1218
Epoch 243/1000
6/6 - 0s - loss: 0.0729 - mae: 0.1218
Epoch 244/1000
6/6 - 0s - loss: 0.0719 - mae: 0.1206
Epoch 245/1000
6/6 - 0s - loss: 0.0724 - mae: 0.1214
Epoch 246/1000
6/6 - 0s - loss: 0.0720 - mae: 0.1212
Epoch 247/1000
6/6 - 0s - loss: 0.0725 - mae: 0.1217
Epoch 248/1000
6/6 - 0s - loss: 0.0724 - mae: 0.1224
Epoch 249/1000
6/6 - 0s - loss: 0.0724 - mae: 0.1218
Epoch 250/1000
6/6 - 0s - loss: 0.0714 - mae: 0.1209
Epoch 251/1000
6/6 - 0s - loss: 0.0719 - mae: 0.1204
Epoch 252/1000
6/6 - 0s - loss: 0.0714 - mae: 0.1204
Epoch 253/1000
6/6 - 0s - loss: 0.0709 - mae: 0.1203
Epoch 254/1000
6/6 - 0s - loss: 0.0709 - mae: 0.1202
Epoch 255/1000
6/6 - 0s - loss: 0.0711 - mae: 0.1204
Epoch 256/1000
6/6 - 0s - loss: 0.0720 - mae: 0.1207
Epoch 257/1000
6/6 - 0s - loss: 0.0720 - mae: 0.1210
Epoch 258/1000
6/6 - 0s - loss: 0.0719 - mae: 0.1227
Epoch 259/1000
6/6 - 0s - loss: 0.0715 - mae: 0.1206
Epoch 260/1000
6/6 - 0s - loss: 0.0713 - mae: 0.1206
Epoch 261/1000
6/6 - 0s - loss: 0.0712 - mae: 0.1200
Epoch 262/1000
6/6 - 0s - loss: 0.0715 - mae: 0.1206
Epoch 263/1000
6/6 - 0s - loss: 0.0711 - mae: 0.1201
Epoch 264/1000
6/6 - 0s - loss: 0.0699 - mae: 0.1195
Epoch 265/1000
6/6 - 0s - loss: 0.0707 - mae: 0.1206
Epoch 266/1000
6/6 - 0s - loss: 0.0710 - mae: 0.1200
Epoch 267/1000
6/6 - 0s - loss: 0.0714 - mae: 0.1195
Epoch 268/1000
6/6 - 0s - loss: 0.0707 - mae: 0.1202
Epoch 269/1000
6/6 - 0s - loss: 0.0702 - mae: 0.1190
Epoch 270/1000
6/6 - 0s - loss: 0.0694 - mae: 0.1185
Epoch 271/1000
6/6 - 0s - loss: 0.0702 - mae: 0.1190
Epoch 272/1000
6/6 - 0s - loss: 0.0704 - mae: 0.1196
Epoch 273/1000
6/6 - 0s - loss: 0.0698 - mae: 0.1189
Epoch 274/1000
6/6 - 0s - loss: 0.0695 - mae: 0.1187
Epoch 275/1000
6/6 - 0s - loss: 0.0699 - mae: 0.1187
Epoch 276/1000
6/6 - 1s - loss: 0.0693 - mae: 0.1190
Epoch 277/1000
6/6 - 0s - loss: 0.0695 - mae: 0.1190
Epoch 278/1000
6/6 - 0s - loss: 0.0691 - mae: 0.1181
Epoch 279/1000
6/6 - 1s - loss: 0.0691 - mae: 0.1180
Epoch 280/1000
6/6 - 0s - loss: 0.0689 - mae: 0.1183
Epoch 281/1000
6/6 - 0s - loss: 0.0699 - mae: 0.1189
Epoch 282/1000
6/6 - 0s - loss: 0.0693 - mae: 0.1175
Epoch 283/1000
6/6 - 0s - loss: 0.0694 - mae: 0.1184
Epoch 284/1000
6/6 - 0s - loss: 0.0702 - mae: 0.1188
Epoch 285/1000
6/6 - 0s - loss: 0.0701 - mae: 0.1192
Epoch 286/1000
6/6 - 0s - loss: 0.0686 - mae: 0.1185
Epoch 287/1000
6/6 - 0s - loss: 0.0689 - mae: 0.1178
Epoch 288/1000
6/6 - 0s - loss: 0.0685 - mae: 0.1183
Epoch 289/1000
6/6 - 0s - loss: 0.0684 - mae: 0.1172
Epoch 290/1000
6/6 - 0s - loss: 0.0690 - mae: 0.1180
Epoch 291/1000
6/6 - 0s - loss: 0.0693 - mae: 0.1175
Epoch 292/1000
6/6 - 0s - loss: 0.0678 - mae: 0.1167
Epoch 293/1000
6/6 - 0s - loss: 0.0675 - mae: 0.1173
Epoch 294/1000
6/6 - 0s - loss: 0.0675 - mae: 0.1165
Epoch 295/1000
6/6 - 0s - loss: 0.0682 - mae: 0.1171
Epoch 296/1000
6/6 - 0s - loss: 0.0676 - mae: 0.1165
Epoch 297/1000
6/6 - 0s - loss: 0.0677 - mae: 0.1167
Epoch 298/1000
6/6 - 0s - loss: 0.0678 - mae: 0.1171
Epoch 299/1000
6/6 - 0s - loss: 0.0676 - mae: 0.1170
Epoch 300/1000
6/6 - 0s - loss: 0.0672 - mae: 0.1165
Epoch 301/1000
6/6 - 0s - loss: 0.0675 - mae: 0.1166
Epoch 302/1000
6/6 - 0s - loss: 0.0674 - mae: 0.1168
Epoch 303/1000
6/6 - 0s - loss: 0.0672 - mae: 0.1162
Epoch 304/1000
6/6 - 0s - loss: 0.0680 - mae: 0.1169
Epoch 305/1000
6/6 - 0s - loss: 0.0685 - mae: 0.1176
Epoch 306/1000
6/6 - 0s - loss: 0.0680 - mae: 0.1170
Epoch 307/1000
6/6 - 0s - loss: 0.0676 - mae: 0.1170
Epoch 308/1000
6/6 - 0s - loss: 0.0661 - mae: 0.1153
Epoch 309/1000
6/6 - 0s - loss: 0.0671 - mae: 0.1161
Epoch 310/1000
6/6 - 0s - loss: 0.0677 - mae: 0.1167
Epoch 311/1000
6/6 - 1s - loss: 0.0674 - mae: 0.1164
Epoch 312/1000
6/6 - 0s - loss: 0.0673 - mae: 0.1161
Epoch 313/1000
6/6 - 0s - loss: 0.0671 - mae: 0.1163
Epoch 314/1000
6/6 - 0s - loss: 0.0666 - mae: 0.1155
Epoch 315/1000
6/6 - 1s - loss: 0.0668 - mae: 0.1163
Epoch 316/1000
6/6 - 0s - loss: 0.0672 - mae: 0.1165
Epoch 317/1000
6/6 - 0s - loss: 0.0666 - mae: 0.1162
Epoch 318/1000
6/6 - 0s - loss: 0.0663 - mae: 0.1156
Epoch 319/1000
6/6 - 0s - loss: 0.0662 - mae: 0.1161
Epoch 320/1000
6/6 - 0s - loss: 0.0665 - mae: 0.1161
Epoch 321/1000
6/6 - 0s - loss: 0.0666 - mae: 0.1159
Epoch 322/1000
6/6 - 0s - loss: 0.0655 - mae: 0.1158
Epoch 323/1000
6/6 - 0s - loss: 0.0656 - mae: 0.1151
Epoch 324/1000
6/6 - 0s - loss: 0.0657 - mae: 0.1157
Epoch 325/1000
6/6 - 0s - loss: 0.0660 - mae: 0.1157
Epoch 326/1000
6/6 - 0s - loss: 0.0660 - mae: 0.1163
Epoch 327/1000
6/6 - 0s - loss: 0.0657 - mae: 0.1155
Epoch 328/1000
6/6 - 0s - loss: 0.0653 - mae: 0.1158
Epoch 329/1000
6/6 - 0s - loss: 0.0660 - mae: 0.1161
Epoch 330/1000
6/6 - 0s - loss: 0.0658 - mae: 0.1162
Epoch 331/1000
6/6 - 1s - loss: 0.0660 - mae: 0.1154
Epoch 332/1000
6/6 - 0s - loss: 0.0655 - mae: 0.1157
Epoch 333/1000
6/6 - 0s - loss: 0.0656 - mae: 0.1156
Epoch 334/1000
6/6 - 0s - loss: 0.0656 - mae: 0.1151
Epoch 335/1000
6/6 - 0s - loss: 0.0656 - mae: 0.1151
Epoch 336/1000
6/6 - 0s - loss: 0.0648 - mae: 0.1153
Epoch 337/1000
6/6 - 0s - loss: 0.0655 - mae: 0.1152
Epoch 338/1000
6/6 - 0s - loss: 0.0648 - mae: 0.1144
Epoch 339/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1144
Epoch 340/1000
6/6 - 0s - loss: 0.0652 - mae: 0.1148
Epoch 341/1000
6/6 - 0s - loss: 0.0647 - mae: 0.1138
Epoch 342/1000
6/6 - 0s - loss: 0.0645 - mae: 0.1142
Epoch 343/1000
6/6 - 0s - loss: 0.0647 - mae: 0.1143
Epoch 344/1000
6/6 - 0s - loss: 0.0646 - mae: 0.1148
Epoch 345/1000
6/6 - 1s - loss: 0.0640 - mae: 0.1133
Epoch 346/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1138
Epoch 347/1000
6/6 - 0s - loss: 0.0639 - mae: 0.1143
Epoch 348/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1150
Epoch 349/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1143
Epoch 350/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1138
Epoch 351/1000
6/6 - 0s - loss: 0.0636 - mae: 0.1134
Epoch 352/1000
6/6 - 0s - loss: 0.0643 - mae: 0.1151
Epoch 353/1000
6/6 - 0s - loss: 0.0636 - mae: 0.1140
Epoch 354/1000
6/6 - 0s - loss: 0.0639 - mae: 0.1139
Epoch 355/1000
6/6 - 0s - loss: 0.0639 - mae: 0.1133
Epoch 356/1000
6/6 - 0s - loss: 0.0637 - mae: 0.1140
Epoch 357/1000
6/6 - 0s - loss: 0.0648 - mae: 0.1148
Epoch 358/1000
6/6 - 0s - loss: 0.0635 - mae: 0.1136
Epoch 359/1000
6/6 - 0s - loss: 0.0635 - mae: 0.1137
Epoch 360/1000
6/6 - 0s - loss: 0.0642 - mae: 0.1133
Epoch 361/1000
6/6 - 0s - loss: 0.0638 - mae: 0.1138
Epoch 362/1000
6/6 - 0s - loss: 0.0639 - mae: 0.1142
Epoch 363/1000
6/6 - 0s - loss: 0.0630 - mae: 0.1137
Epoch 364/1000
6/6 - 0s - loss: 0.0628 - mae: 0.1129
Epoch 365/1000
6/6 - 1s - loss: 0.0639 - mae: 0.1143
Epoch 366/1000
6/6 - 0s - loss: 0.0640 - mae: 0.1148
Epoch 367/1000
6/6 - 0s - loss: 0.0631 - mae: 0.1135
Epoch 368/1000
6/6 - 0s - loss: 0.0632 - mae: 0.1134
Epoch 369/1000
6/6 - 0s - loss: 0.0635 - mae: 0.1136
Epoch 370/1000
6/6 - 0s - loss: 0.0622 - mae: 0.1124
Epoch 371/1000
6/6 - 0s - loss: 0.0621 - mae: 0.1119
Epoch 372/1000
6/6 - 0s - loss: 0.0627 - mae: 0.1128
Epoch 373/1000
6/6 - 0s - loss: 0.0624 - mae: 0.1127
Epoch 374/1000
6/6 - 1s - loss: 0.0626 - mae: 0.1125
Epoch 375/1000
6/6 - 0s - loss: 0.0625 - mae: 0.1127
Epoch 376/1000
6/6 - 0s - loss: 0.0623 - mae: 0.1124
Epoch 377/1000
6/6 - 0s - loss: 0.0630 - mae: 0.1133
Epoch 378/1000
6/6 - 0s - loss: 0.0626 - mae: 0.1129
Epoch 379/1000
6/6 - 0s - loss: 0.0626 - mae: 0.1131
Epoch 380/1000
6/6 - 0s - loss: 0.0626 - mae: 0.1132
Epoch 381/1000
6/6 - 0s - loss: 0.0627 - mae: 0.1139
Epoch 382/1000
6/6 - 0s - loss: 0.0627 - mae: 0.1127
Epoch 383/1000
6/6 - 0s - loss: 0.0629 - mae: 0.1129
Epoch 384/1000
6/6 - 0s - loss: 0.0620 - mae: 0.1123
Epoch 385/1000
6/6 - 0s - loss: 0.0620 - mae: 0.1119
Epoch 386/1000
6/6 - 0s - loss: 0.0619 - mae: 0.1123
Epoch 387/1000
6/6 - 0s - loss: 0.0613 - mae: 0.1125
Epoch 388/1000
6/6 - 0s - loss: 0.0612 - mae: 0.1118
Epoch 389/1000
6/6 - 0s - loss: 0.0614 - mae: 0.1118
Epoch 390/1000
6/6 - 0s - loss: 0.0612 - mae: 0.1113
Epoch 391/1000
6/6 - 0s - loss: 0.0613 - mae: 0.1112
Epoch 392/1000
6/6 - 0s - loss: 0.0616 - mae: 0.1126
Epoch 393/1000
6/6 - 0s - loss: 0.0613 - mae: 0.1118
Epoch 394/1000
6/6 - 0s - loss: 0.0616 - mae: 0.1115
Epoch 395/1000
6/6 - 0s - loss: 0.0612 - mae: 0.1118
Epoch 396/1000
6/6 - 1s - loss: 0.0608 - mae: 0.1121
Epoch 397/1000
6/6 - 0s - loss: 0.0612 - mae: 0.1123
Epoch 398/1000
6/6 - 0s - loss: 0.0611 - mae: 0.1113
Epoch 399/1000
6/6 - 0s - loss: 0.0609 - mae: 0.1112
Epoch 400/1000
6/6 - 1s - loss: 0.0612 - mae: 0.1115
Epoch 401/1000
6/6 - 0s - loss: 0.0605 - mae: 0.1111
Epoch 402/1000
6/6 - 1s - loss: 0.0606 - mae: 0.1117
Epoch 403/1000
6/6 - 1s - loss: 0.0609 - mae: 0.1113
Epoch 404/1000
6/6 - 0s - loss: 0.0608 - mae: 0.1118
Epoch 405/1000
6/6 - 1s - loss: 0.0604 - mae: 0.1110
Epoch 406/1000
6/6 - 1s - loss: 0.0603 - mae: 0.1110
Epoch 407/1000
6/6 - 0s - loss: 0.0602 - mae: 0.1111
Epoch 408/1000
6/6 - 0s - loss: 0.0601 - mae: 0.1100
Epoch 409/1000
6/6 - 0s - loss: 0.0605 - mae: 0.1114
Epoch 410/1000
6/6 - 0s - loss: 0.0605 - mae: 0.1120
Epoch 411/1000
6/6 - 0s - loss: 0.0598 - mae: 0.1103
Epoch 412/1000
6/6 - 0s - loss: 0.0604 - mae: 0.1111
Epoch 413/1000
6/6 - 0s - loss: 0.0599 - mae: 0.1110
Epoch 414/1000
6/6 - 0s - loss: 0.0593 - mae: 0.1100
Epoch 415/1000
6/6 - 0s - loss: 0.0601 - mae: 0.1111
Epoch 416/1000
6/6 - 0s - loss: 0.0600 - mae: 0.1107
Epoch 417/1000
6/6 - 0s - loss: 0.0599 - mae: 0.1103
Epoch 418/1000
6/6 - 0s - loss: 0.0593 - mae: 0.1099
Epoch 419/1000
6/6 - 0s - loss: 0.0595 - mae: 0.1100
Epoch 420/1000
6/6 - 0s - loss: 0.0596 - mae: 0.1099
Epoch 421/1000
6/6 - 0s - loss: 0.0591 - mae: 0.1096
Epoch 422/1000
6/6 - 0s - loss: 0.0599 - mae: 0.1107
Epoch 423/1000
6/6 - 0s - loss: 0.0596 - mae: 0.1100
Epoch 424/1000
6/6 - 0s - loss: 0.0588 - mae: 0.1097
Epoch 425/1000
6/6 - 0s - loss: 0.0594 - mae: 0.1101
Epoch 426/1000
6/6 - 0s - loss: 0.0597 - mae: 0.1111
Epoch 427/1000
6/6 - 1s - loss: 0.0594 - mae: 0.1103
Epoch 428/1000
6/6 - 0s - loss: 0.0588 - mae: 0.1098
Epoch 429/1000
6/6 - 0s - loss: 0.0592 - mae: 0.1096
Epoch 430/1000
6/6 - 1s - loss: 0.0591 - mae: 0.1099
Epoch 431/1000
6/6 - 0s - loss: 0.0589 - mae: 0.1100
Epoch 432/1000
6/6 - 0s - loss: 0.0597 - mae: 0.1103
Epoch 433/1000
6/6 - 1s - loss: 0.0597 - mae: 0.1118
Epoch 434/1000
6/6 - 1s - loss: 0.0592 - mae: 0.1104
Epoch 435/1000
6/6 - 0s - loss: 0.0592 - mae: 0.1101
Epoch 436/1000
6/6 - 0s - loss: 0.0589 - mae: 0.1100
Epoch 437/1000
6/6 - 0s - loss: 0.0588 - mae: 0.1103
Epoch 438/1000
6/6 - 0s - loss: 0.0587 - mae: 0.1090
Epoch 439/1000
6/6 - 0s - loss: 0.0582 - mae: 0.1095
Epoch 440/1000
6/6 - 0s - loss: 0.0582 - mae: 0.1092
Epoch 441/1000
6/6 - 0s - loss: 0.0580 - mae: 0.1095
Epoch 442/1000
6/6 - 0s - loss: 0.0573 - mae: 0.1088
Epoch 443/1000
6/6 - 0s - loss: 0.0584 - mae: 0.1094
Epoch 444/1000
6/6 - 1s - loss: 0.0576 - mae: 0.1091
Epoch 445/1000
6/6 - 0s - loss: 0.0585 - mae: 0.1098
Epoch 446/1000
6/6 - 0s - loss: 0.0576 - mae: 0.1083
Epoch 447/1000
6/6 - 0s - loss: 0.0581 - mae: 0.1094
Epoch 448/1000
6/6 - 0s - loss: 0.0577 - mae: 0.1093
Epoch 449/1000
6/6 - 0s - loss: 0.0580 - mae: 0.1093
Epoch 450/1000
6/6 - 0s - loss: 0.0579 - mae: 0.1090
Epoch 451/1000
6/6 - 0s - loss: 0.0582 - mae: 0.1092
Epoch 452/1000
6/6 - 0s - loss: 0.0593 - mae: 0.1099
Epoch 453/1000
6/6 - 0s - loss: 0.0592 - mae: 0.1106
Epoch 454/1000
6/6 - 0s - loss: 0.0592 - mae: 0.1118
Epoch 455/1000
6/6 - 0s - loss: 0.0591 - mae: 0.1108
Epoch 456/1000
6/6 - 0s - loss: 0.0584 - mae: 0.1101
Epoch 457/1000
6/6 - 0s - loss: 0.0579 - mae: 0.1103
Epoch 458/1000
6/6 - 0s - loss: 0.0579 - mae: 0.1096
Epoch 459/1000
6/6 - 0s - loss: 0.0568 - mae: 0.1088
Epoch 460/1000
6/6 - 0s - loss: 0.0574 - mae: 0.1084
Epoch 461/1000
6/6 - 1s - loss: 0.0576 - mae: 0.1084
Epoch 462/1000
6/6 - 0s - loss: 0.0569 - mae: 0.1085
Epoch 463/1000
6/6 - 0s - loss: 0.0566 - mae: 0.1079
Epoch 464/1000
6/6 - 0s - loss: 0.0566 - mae: 0.1079
Epoch 465/1000
6/6 - 1s - loss: 0.0562 - mae: 0.1074
Epoch 466/1000
6/6 - 0s - loss: 0.0564 - mae: 0.1080
Epoch 467/1000
6/6 - 0s - loss: 0.0569 - mae: 0.1079
Epoch 468/1000
6/6 - 0s - loss: 0.0569 - mae: 0.1086
Epoch 469/1000
6/6 - 0s - loss: 0.0562 - mae: 0.1089
Epoch 470/1000
6/6 - 0s - loss: 0.0562 - mae: 0.1083
Epoch 471/1000
6/6 - 0s - loss: 0.0561 - mae: 0.1074
Epoch 472/1000
6/6 - 0s - loss: 0.0563 - mae: 0.1078
Epoch 473/1000
6/6 - 0s - loss: 0.0563 - mae: 0.1077
Epoch 474/1000
6/6 - 0s - loss: 0.0556 - mae: 0.1075
Epoch 475/1000
6/6 - 0s - loss: 0.0563 - mae: 0.1084
Epoch 476/1000
6/6 - 0s - loss: 0.0567 - mae: 0.1081
Epoch 477/1000
6/6 - 0s - loss: 0.0560 - mae: 0.1083
Epoch 478/1000
6/6 - 0s - loss: 0.0560 - mae: 0.1081
Epoch 479/1000
6/6 - 0s - loss: 0.0557 - mae: 0.1074
Epoch 480/1000
6/6 - 0s - loss: 0.0558 - mae: 0.1075
Epoch 481/1000
6/6 - 0s - loss: 0.0553 - mae: 0.1074
Epoch 482/1000
6/6 - 0s - loss: 0.0558 - mae: 0.1078
Epoch 483/1000
6/6 - 0s - loss: 0.0552 - mae: 0.1071
Epoch 484/1000
6/6 - 0s - loss: 0.0549 - mae: 0.1068
Epoch 485/1000
6/6 - 0s - loss: 0.0552 - mae: 0.1068
Epoch 486/1000
6/6 - 0s - loss: 0.0551 - mae: 0.1069
Epoch 487/1000
6/6 - 0s - loss: 0.0554 - mae: 0.1072
Epoch 488/1000
6/6 - 0s - loss: 0.0553 - mae: 0.1073
Epoch 489/1000
6/6 - 0s - loss: 0.0549 - mae: 0.1070
Epoch 490/1000
6/6 - 0s - loss: 0.0544 - mae: 0.1067
Epoch 491/1000
6/6 - 0s - loss: 0.0549 - mae: 0.1071
Epoch 492/1000
6/6 - 0s - loss: 0.0551 - mae: 0.1066
Epoch 493/1000
6/6 - 0s - loss: 0.0545 - mae: 0.1067
Epoch 494/1000
6/6 - 0s - loss: 0.0547 - mae: 0.1071
Epoch 495/1000
6/6 - 0s - loss: 0.0543 - mae: 0.1061
Epoch 496/1000
6/6 - 1s - loss: 0.0543 - mae: 0.1066
Epoch 497/1000
6/6 - 0s - loss: 0.0541 - mae: 0.1066
Epoch 498/1000
6/6 - 0s - loss: 0.0545 - mae: 0.1062
Epoch 499/1000
6/6 - 0s - loss: 0.0540 - mae: 0.1068
Epoch 500/1000
6/6 - 0s - loss: 0.0544 - mae: 0.1067
Epoch 501/1000
6/6 - 0s - loss: 0.0537 - mae: 0.1057
Epoch 502/1000
6/6 - 1s - loss: 0.0537 - mae: 0.1060
Epoch 503/1000
6/6 - 0s - loss: 0.0545 - mae: 0.1070
Epoch 504/1000
6/6 - 0s - loss: 0.0541 - mae: 0.1069
Epoch 505/1000
6/6 - 0s - loss: 0.0537 - mae: 0.1066
Epoch 506/1000
6/6 - 0s - loss: 0.0542 - mae: 0.1069
Epoch 507/1000
6/6 - 0s - loss: 0.0543 - mae: 0.1073
Epoch 508/1000
6/6 - 0s - loss: 0.0536 - mae: 0.1069
Epoch 509/1000
6/6 - 0s - loss: 0.0533 - mae: 0.1059
Epoch 510/1000
6/6 - 0s - loss: 0.0533 - mae: 0.1062
Epoch 511/1000
6/6 - 0s - loss: 0.0532 - mae: 0.1059
Epoch 512/1000
6/6 - 0s - loss: 0.0526 - mae: 0.1053
Epoch 513/1000
6/6 - 0s - loss: 0.0526 - mae: 0.1059
Epoch 514/1000
6/6 - 0s - loss: 0.0529 - mae: 0.1056
Epoch 515/1000
6/6 - 0s - loss: 0.0525 - mae: 0.1058
Epoch 516/1000
6/6 - 0s - loss: 0.0527 - mae: 0.1051
Epoch 517/1000
6/6 - 0s - loss: 0.0532 - mae: 0.1061
Epoch 518/1000
6/6 - 0s - loss: 0.0524 - mae: 0.1049
Epoch 519/1000
6/6 - 0s - loss: 0.0525 - mae: 0.1053
Epoch 520/1000
6/6 - 0s - loss: 0.0525 - mae: 0.1052
Epoch 521/1000
6/6 - 1s - loss: 0.0530 - mae: 0.1054
Epoch 522/1000
6/6 - 1s - loss: 0.0529 - mae: 0.1058
Epoch 523/1000
6/6 - 0s - loss: 0.0526 - mae: 0.1059
Epoch 524/1000
6/6 - 0s - loss: 0.0526 - mae: 0.1057
Epoch 525/1000
6/6 - 0s - loss: 0.0524 - mae: 0.1061
Epoch 526/1000
6/6 - 0s - loss: 0.0521 - mae: 0.1055
Epoch 527/1000
6/6 - 0s - loss: 0.0521 - mae: 0.1047
Epoch 528/1000
6/6 - 0s - loss: 0.0519 - mae: 0.1046
Epoch 529/1000
6/6 - 0s - loss: 0.0521 - mae: 0.1049
Epoch 530/1000
6/6 - 0s - loss: 0.0519 - mae: 0.1051
Epoch 531/1000
6/6 - 0s - loss: 0.0521 - mae: 0.1059
Epoch 532/1000
6/6 - 0s - loss: 0.0519 - mae: 0.1051
Epoch 533/1000
6/6 - 1s - loss: 0.0521 - mae: 0.1051
Epoch 534/1000
6/6 - 0s - loss: 0.0520 - mae: 0.1057
Epoch 535/1000
6/6 - 0s - loss: 0.0515 - mae: 0.1058
Epoch 536/1000
6/6 - 0s - loss: 0.0511 - mae: 0.1054
Epoch 537/1000
6/6 - 1s - loss: 0.0518 - mae: 0.1051
Epoch 538/1000
6/6 - 0s - loss: 0.0515 - mae: 0.1051
Epoch 539/1000
6/6 - 0s - loss: 0.0511 - mae: 0.1050
Epoch 540/1000
6/6 - 0s - loss: 0.0509 - mae: 0.1041
Epoch 541/1000
6/6 - 0s - loss: 0.0514 - mae: 0.1045
Epoch 542/1000
6/6 - 0s - loss: 0.0509 - mae: 0.1041
Epoch 543/1000
6/6 - 1s - loss: 0.0507 - mae: 0.1041
Epoch 544/1000
6/6 - 0s - loss: 0.0508 - mae: 0.1038
Epoch 545/1000
6/6 - 1s - loss: 0.0507 - mae: 0.1044
Epoch 546/1000
6/6 - 0s - loss: 0.0505 - mae: 0.1041
Epoch 547/1000
6/6 - 0s - loss: 0.0508 - mae: 0.1043
Epoch 548/1000
6/6 - 0s - loss: 0.0501 - mae: 0.1045
Epoch 549/1000
6/6 - 0s - loss: 0.0498 - mae: 0.1038
Epoch 550/1000
6/6 - 0s - loss: 0.0501 - mae: 0.1040
Epoch 551/1000
6/6 - 0s - loss: 0.0498 - mae: 0.1040
Epoch 552/1000
6/6 - 0s - loss: 0.0495 - mae: 0.1032
Epoch 553/1000
6/6 - 0s - loss: 0.0499 - mae: 0.1035
Epoch 554/1000
6/6 - 0s - loss: 0.0499 - mae: 0.1035
Epoch 555/1000
6/6 - 0s - loss: 0.0500 - mae: 0.1042
Epoch 556/1000
6/6 - 0s - loss: 0.0495 - mae: 0.1033
Epoch 557/1000
6/6 - 0s - loss: 0.0497 - mae: 0.1035
Epoch 558/1000
6/6 - 0s - loss: 0.0497 - mae: 0.1035
Epoch 559/1000
6/6 - 0s - loss: 0.0491 - mae: 0.1033
Epoch 560/1000
6/6 - 0s - loss: 0.0496 - mae: 0.1039
Epoch 561/1000
6/6 - 0s - loss: 0.0498 - mae: 0.1044
Epoch 562/1000
6/6 - 0s - loss: 0.0501 - mae: 0.1055
Epoch 563/1000
6/6 - 0s - loss: 0.0492 - mae: 0.1044
Epoch 564/1000
6/6 - 0s - loss: 0.0495 - mae: 0.1040
Epoch 565/1000
6/6 - 0s - loss: 0.0491 - mae: 0.1034
Epoch 566/1000
6/6 - 0s - loss: 0.0487 - mae: 0.1028
Epoch 567/1000
6/6 - 0s - loss: 0.0489 - mae: 0.1030
Epoch 568/1000
6/6 - 0s - loss: 0.0488 - mae: 0.1030
Epoch 569/1000
6/6 - 0s - loss: 0.0487 - mae: 0.1036
Epoch 570/1000
6/6 - 0s - loss: 0.0488 - mae: 0.1032
Epoch 571/1000
6/6 - 0s - loss: 0.0482 - mae: 0.1023
Epoch 572/1000
6/6 - 0s - loss: 0.0485 - mae: 0.1026
Epoch 573/1000
6/6 - 0s - loss: 0.0487 - mae: 0.1027
Epoch 574/1000
6/6 - 1s - loss: 0.0478 - mae: 0.1024
Epoch 575/1000
6/6 - 0s - loss: 0.0479 - mae: 0.1023
Epoch 576/1000
6/6 - 1s - loss: 0.0483 - mae: 0.1028
Epoch 577/1000
6/6 - 0s - loss: 0.0477 - mae: 0.1026
Epoch 578/1000
6/6 - 0s - loss: 0.0476 - mae: 0.1022
Epoch 579/1000
6/6 - 1s - loss: 0.0478 - mae: 0.1025
Epoch 580/1000
6/6 - 0s - loss: 0.0474 - mae: 0.1020
Epoch 581/1000
6/6 - 0s - loss: 0.0477 - mae: 0.1026
Epoch 582/1000
6/6 - 0s - loss: 0.0477 - mae: 0.1021
Epoch 583/1000
6/6 - 0s - loss: 0.0478 - mae: 0.1023
Epoch 584/1000
6/6 - 0s - loss: 0.0473 - mae: 0.1019
Epoch 585/1000
6/6 - 0s - loss: 0.0474 - mae: 0.1026
Epoch 586/1000
6/6 - 0s - loss: 0.0470 - mae: 0.1021
Epoch 587/1000
6/6 - 0s - loss: 0.0472 - mae: 0.1022
Epoch 588/1000
6/6 - 0s - loss: 0.0464 - mae: 0.1012
Epoch 589/1000
6/6 - 0s - loss: 0.0464 - mae: 0.1015
Epoch 590/1000
6/6 - 0s - loss: 0.0466 - mae: 0.1015
Epoch 591/1000
6/6 - 0s - loss: 0.0466 - mae: 0.1019
Epoch 592/1000
6/6 - 0s - loss: 0.0464 - mae: 0.1020
Epoch 593/1000
6/6 - 0s - loss: 0.0466 - mae: 0.1016
Epoch 594/1000
6/6 - 0s - loss: 0.0464 - mae: 0.1012
Epoch 595/1000
6/6 - 0s - loss: 0.0471 - mae: 0.1027
Epoch 596/1000
6/6 - 0s - loss: 0.0468 - mae: 0.1036
Epoch 597/1000
6/6 - 0s - loss: 0.0469 - mae: 0.1020
Epoch 598/1000
6/6 - 0s - loss: 0.0468 - mae: 0.1026
Epoch 599/1000
6/6 - 0s - loss: 0.0464 - mae: 0.1014
Epoch 600/1000
6/6 - 0s - loss: 0.0463 - mae: 0.1015
Epoch 601/1000
6/6 - 0s - loss: 0.0463 - mae: 0.1015
Epoch 602/1000
6/6 - 0s - loss: 0.0457 - mae: 0.1005
Epoch 603/1000
6/6 - 0s - loss: 0.0454 - mae: 0.1011
Epoch 604/1000
6/6 - 0s - loss: 0.0455 - mae: 0.1013
Epoch 605/1000
6/6 - 0s - loss: 0.0455 - mae: 0.1009
Epoch 606/1000
6/6 - 0s - loss: 0.0453 - mae: 0.1006
Epoch 607/1000
6/6 - 0s - loss: 0.0455 - mae: 0.1008
Epoch 608/1000
6/6 - 1s - loss: 0.0447 - mae: 0.1004
Epoch 609/1000
6/6 - 0s - loss: 0.0451 - mae: 0.1008
Epoch 610/1000
6/6 - 0s - loss: 0.0452 - mae: 0.1004
Epoch 611/1000
6/6 - 0s - loss: 0.0456 - mae: 0.1010
Epoch 612/1000
6/6 - 1s - loss: 0.0449 - mae: 0.1004
Epoch 613/1000
6/6 - 0s - loss: 0.0451 - mae: 0.1007
Epoch 614/1000
6/6 - 0s - loss: 0.0454 - mae: 0.1017
Epoch 615/1000
6/6 - 1s - loss: 0.0451 - mae: 0.1024
Epoch 616/1000
6/6 - 0s - loss: 0.0455 - mae: 0.1017
Epoch 617/1000
6/6 - 0s - loss: 0.0451 - mae: 0.1007
Epoch 618/1000
6/6 - 0s - loss: 0.0446 - mae: 0.1003
Epoch 619/1000
6/6 - 0s - loss: 0.0448 - mae: 0.1010
Epoch 620/1000
6/6 - 0s - loss: 0.0444 - mae: 0.1007
Epoch 621/1000
6/6 - 0s - loss: 0.0443 - mae: 0.1005
Epoch 622/1000
6/6 - 0s - loss: 0.0444 - mae: 0.1003
Epoch 623/1000
6/6 - 0s - loss: 0.0445 - mae: 0.1007
Epoch 624/1000
6/6 - 1s - loss: 0.0444 - mae: 0.1003
Epoch 625/1000
6/6 - 1s - loss: 0.0445 - mae: 0.0998
Epoch 626/1000
6/6 - 0s - loss: 0.0439 - mae: 0.1000
Epoch 627/1000
6/6 - 0s - loss: 0.0445 - mae: 0.1005
Epoch 628/1000
6/6 - 0s - loss: 0.0437 - mae: 0.1002
Epoch 629/1000
6/6 - 0s - loss: 0.0440 - mae: 0.0993
Epoch 630/1000
6/6 - 0s - loss: 0.0433 - mae: 0.0993
Epoch 631/1000
6/6 - 0s - loss: 0.0437 - mae: 0.0996
Epoch 632/1000
6/6 - 0s - loss: 0.0436 - mae: 0.1000
Epoch 633/1000
6/6 - 0s - loss: 0.0437 - mae: 0.0996
Epoch 634/1000
6/6 - 0s - loss: 0.0434 - mae: 0.0993
Epoch 635/1000
6/6 - 0s - loss: 0.0435 - mae: 0.1001
Epoch 636/1000
6/6 - 0s - loss: 0.0432 - mae: 0.0997
Epoch 637/1000
6/6 - 0s - loss: 0.0427 - mae: 0.0993
Epoch 638/1000
6/6 - 0s - loss: 0.0429 - mae: 0.0993
Epoch 639/1000
6/6 - 0s - loss: 0.0431 - mae: 0.0996
Epoch 640/1000
6/6 - 0s - loss: 0.0432 - mae: 0.0999
Epoch 641/1000
6/6 - 1s - loss: 0.0430 - mae: 0.0993
Epoch 642/1000
6/6 - 0s - loss: 0.0428 - mae: 0.0994
Epoch 643/1000
6/6 - 0s - loss: 0.0428 - mae: 0.0990
Epoch 644/1000
6/6 - 0s - loss: 0.0429 - mae: 0.0997
Epoch 645/1000
6/6 - 0s - loss: 0.0429 - mae: 0.1003
Epoch 646/1000
6/6 - 1s - loss: 0.0430 - mae: 0.1001
Epoch 647/1000
6/6 - 0s - loss: 0.0429 - mae: 0.1008
Epoch 648/1000
6/6 - 0s - loss: 0.0425 - mae: 0.0999
Epoch 649/1000
6/6 - 0s - loss: 0.0424 - mae: 0.0992
Epoch 650/1000
6/6 - 0s - loss: 0.0421 - mae: 0.0987
Epoch 651/1000
6/6 - 0s - loss: 0.0423 - mae: 0.0996
Epoch 652/1000
6/6 - 0s - loss: 0.0421 - mae: 0.0995
Epoch 653/1000
6/6 - 0s - loss: 0.0421 - mae: 0.0993
Epoch 654/1000
6/6 - 0s - loss: 0.0418 - mae: 0.0983
Epoch 655/1000
6/6 - 0s - loss: 0.0416 - mae: 0.0984
Epoch 656/1000
6/6 - 0s - loss: 0.0424 - mae: 0.0996
Epoch 657/1000
6/6 - 0s - loss: 0.0417 - mae: 0.0986
Epoch 658/1000
6/6 - 1s - loss: 0.0416 - mae: 0.0990
Epoch 659/1000
6/6 - 0s - loss: 0.0417 - mae: 0.0984
Epoch 660/1000
6/6 - 0s - loss: 0.0416 - mae: 0.0991
Epoch 661/1000
6/6 - 0s - loss: 0.0418 - mae: 0.0984
Epoch 662/1000
6/6 - 0s - loss: 0.0415 - mae: 0.0981
Epoch 663/1000
6/6 - 0s - loss: 0.0411 - mae: 0.0978
Epoch 664/1000
6/6 - 0s - loss: 0.0415 - mae: 0.0985
Epoch 665/1000
6/6 - 0s - loss: 0.0410 - mae: 0.0979
Epoch 666/1000
6/6 - 0s - loss: 0.0413 - mae: 0.0979
Epoch 667/1000
6/6 - 1s - loss: 0.0410 - mae: 0.0981
Epoch 668/1000
6/6 - 0s - loss: 0.0413 - mae: 0.0981
Epoch 669/1000
6/6 - 0s - loss: 0.0408 - mae: 0.0977
Epoch 670/1000
6/6 - 0s - loss: 0.0411 - mae: 0.0986
Epoch 671/1000
6/6 - 0s - loss: 0.0409 - mae: 0.0978
Epoch 672/1000
6/6 - 0s - loss: 0.0409 - mae: 0.0978
Epoch 673/1000
6/6 - 0s - loss: 0.0407 - mae: 0.0982
Epoch 674/1000
6/6 - 0s - loss: 0.0403 - mae: 0.0977
Epoch 675/1000
6/6 - 1s - loss: 0.0410 - mae: 0.0982
Epoch 676/1000
6/6 - 0s - loss: 0.0409 - mae: 0.0979
Epoch 677/1000
6/6 - 0s - loss: 0.0402 - mae: 0.0975
Epoch 678/1000
6/6 - 0s - loss: 0.0406 - mae: 0.0978
Epoch 679/1000
6/6 - 0s - loss: 0.0404 - mae: 0.0976
Epoch 680/1000
6/6 - 0s - loss: 0.0407 - mae: 0.0981
Epoch 681/1000
6/6 - 0s - loss: 0.0405 - mae: 0.0979
Epoch 682/1000
6/6 - 0s - loss: 0.0400 - mae: 0.0976
Epoch 683/1000
6/6 - 0s - loss: 0.0407 - mae: 0.0979
Epoch 684/1000
6/6 - 0s - loss: 0.0395 - mae: 0.0974
Epoch 685/1000
6/6 - 0s - loss: 0.0398 - mae: 0.0979
Epoch 686/1000
6/6 - 0s - loss: 0.0401 - mae: 0.0974
Epoch 687/1000
6/6 - 1s - loss: 0.0400 - mae: 0.0968
Epoch 688/1000
6/6 - 0s - loss: 0.0399 - mae: 0.0974
Epoch 689/1000
6/6 - 1s - loss: 0.0399 - mae: 0.0971
Epoch 690/1000
6/6 - 0s - loss: 0.0401 - mae: 0.0973
Epoch 691/1000
6/6 - 0s - loss: 0.0404 - mae: 0.0974
Epoch 692/1000
6/6 - 0s - loss: 0.0397 - mae: 0.0970
Epoch 693/1000
6/6 - 0s - loss: 0.0400 - mae: 0.0978
Epoch 694/1000
6/6 - 0s - loss: 0.0399 - mae: 0.0974
Epoch 695/1000
6/6 - 0s - loss: 0.0397 - mae: 0.0975
Epoch 696/1000
6/6 - 0s - loss: 0.0395 - mae: 0.0975
Epoch 697/1000
6/6 - 0s - loss: 0.0399 - mae: 0.0981
Epoch 698/1000
6/6 - 0s - loss: 0.0400 - mae: 0.0978
Epoch 699/1000
6/6 - 1s - loss: 0.0391 - mae: 0.0964
Epoch 700/1000
6/6 - 0s - loss: 0.0392 - mae: 0.0970
Epoch 701/1000
6/6 - 0s - loss: 0.0389 - mae: 0.0969
Epoch 702/1000
6/6 - 0s - loss: 0.0389 - mae: 0.0966
Epoch 703/1000
6/6 - 0s - loss: 0.0389 - mae: 0.0964
Epoch 704/1000
6/6 - 0s - loss: 0.0390 - mae: 0.0968
Epoch 705/1000
6/6 - 0s - loss: 0.0383 - mae: 0.0962
Epoch 706/1000
6/6 - 1s - loss: 0.0384 - mae: 0.0964
Epoch 707/1000
6/6 - 0s - loss: 0.0383 - mae: 0.0962
Epoch 708/1000
6/6 - 0s - loss: 0.0385 - mae: 0.0962
Epoch 709/1000
6/6 - 1s - loss: 0.0394 - mae: 0.0983
Epoch 710/1000
6/6 - 0s - loss: 0.0385 - mae: 0.0969
Epoch 711/1000
6/6 - 0s - loss: 0.0381 - mae: 0.0959
Epoch 712/1000
6/6 - 1s - loss: 0.0386 - mae: 0.0962
Epoch 713/1000
6/6 - 1s - loss: 0.0382 - mae: 0.0957
Epoch 714/1000
6/6 - 1s - loss: 0.0382 - mae: 0.0961
Epoch 715/1000
6/6 - 0s - loss: 0.0380 - mae: 0.0961
Epoch 716/1000
6/6 - 1s - loss: 0.0381 - mae: 0.0959
Epoch 717/1000
6/6 - 0s - loss: 0.0382 - mae: 0.0958
Epoch 718/1000
6/6 - 0s - loss: 0.0382 - mae: 0.0974
Epoch 719/1000
6/6 - 1s - loss: 0.0380 - mae: 0.0966
Epoch 720/1000
6/6 - 0s - loss: 0.0381 - mae: 0.0963
Epoch 721/1000
6/6 - 1s - loss: 0.0380 - mae: 0.0956
Epoch 722/1000
6/6 - 0s - loss: 0.0380 - mae: 0.0958
Epoch 723/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0958
Epoch 724/1000
6/6 - 0s - loss: 0.0377 - mae: 0.0966
Epoch 725/1000
6/6 - 0s - loss: 0.0379 - mae: 0.0958
Epoch 726/1000
6/6 - 1s - loss: 0.0375 - mae: 0.0960
Epoch 727/1000
6/6 - 0s - loss: 0.0371 - mae: 0.0959
Epoch 728/1000
6/6 - 0s - loss: 0.0371 - mae: 0.0951
Epoch 729/1000
6/6 - 1s - loss: 0.0372 - mae: 0.0949
Epoch 730/1000
6/6 - 0s - loss: 0.0375 - mae: 0.0955
Epoch 731/1000
6/6 - 0s - loss: 0.0376 - mae: 0.0953
Epoch 732/1000
6/6 - 0s - loss: 0.0372 - mae: 0.0957
Epoch 733/1000
6/6 - 0s - loss: 0.0373 - mae: 0.0958
Epoch 734/1000
6/6 - 0s - loss: 0.0372 - mae: 0.0961
Epoch 735/1000
6/6 - 0s - loss: 0.0367 - mae: 0.0952
Epoch 736/1000
6/6 - 0s - loss: 0.0371 - mae: 0.0952
Epoch 737/1000
6/6 - 0s - loss: 0.0369 - mae: 0.0952
Epoch 738/1000
6/6 - 0s - loss: 0.0365 - mae: 0.0951
Epoch 739/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0948
Epoch 740/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0950
Epoch 741/1000
6/6 - 0s - loss: 0.0363 - mae: 0.0942
Epoch 742/1000
6/6 - 0s - loss: 0.0363 - mae: 0.0943
Epoch 743/1000
6/6 - 1s - loss: 0.0370 - mae: 0.0953
Epoch 744/1000
6/6 - 0s - loss: 0.0366 - mae: 0.0951
Epoch 745/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0947
Epoch 746/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0951
Epoch 747/1000
6/6 - 1s - loss: 0.0369 - mae: 0.0953
Epoch 748/1000
6/6 - 0s - loss: 0.0366 - mae: 0.0954
Epoch 749/1000
6/6 - 0s - loss: 0.0365 - mae: 0.0953
Epoch 750/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0954
Epoch 751/1000
6/6 - 0s - loss: 0.0357 - mae: 0.0944
Epoch 752/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0953
Epoch 753/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0949
Epoch 754/1000
6/6 - 1s - loss: 0.0363 - mae: 0.0956
Epoch 755/1000
6/6 - 0s - loss: 0.0363 - mae: 0.0954
Epoch 756/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0950
Epoch 757/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0955
Epoch 758/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0954
Epoch 759/1000
6/6 - 0s - loss: 0.0359 - mae: 0.0954
Epoch 760/1000
6/6 - 0s - loss: 0.0356 - mae: 0.0943
Epoch 761/1000
6/6 - 0s - loss: 0.0362 - mae: 0.0947
Epoch 762/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0944
Epoch 763/1000
6/6 - 0s - loss: 0.0360 - mae: 0.0949
Epoch 764/1000
6/6 - 1s - loss: 0.0353 - mae: 0.0951
Epoch 765/1000
6/6 - 0s - loss: 0.0358 - mae: 0.0952
Epoch 766/1000
6/6 - 1s - loss: 0.0355 - mae: 0.0947
Epoch 767/1000
6/6 - 0s - loss: 0.0353 - mae: 0.0945
Epoch 768/1000
6/6 - 0s - loss: 0.0354 - mae: 0.0945
Epoch 769/1000
6/6 - 0s - loss: 0.0352 - mae: 0.0944
Epoch 770/1000
6/6 - 0s - loss: 0.0364 - mae: 0.0951
Epoch 771/1000
6/6 - 0s - loss: 0.0354 - mae: 0.0941
Epoch 772/1000
6/6 - 0s - loss: 0.0349 - mae: 0.0939
Epoch 773/1000
6/6 - 0s - loss: 0.0354 - mae: 0.0941
Epoch 774/1000
6/6 - 1s - loss: 0.0352 - mae: 0.0945
Epoch 775/1000
6/6 - 0s - loss: 0.0352 - mae: 0.0935
Epoch 776/1000
6/6 - 0s - loss: 0.0345 - mae: 0.0930
Epoch 777/1000
6/6 - 0s - loss: 0.0346 - mae: 0.0937
Epoch 778/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0935
Epoch 779/1000
6/6 - 0s - loss: 0.0350 - mae: 0.0938
Epoch 780/1000
6/6 - 0s - loss: 0.0344 - mae: 0.0937
Epoch 781/1000
6/6 - 1s - loss: 0.0343 - mae: 0.0933
Epoch 782/1000
6/6 - 0s - loss: 0.0347 - mae: 0.0936
Epoch 783/1000
6/6 - 0s - loss: 0.0346 - mae: 0.0936
Epoch 784/1000
6/6 - 0s - loss: 0.0344 - mae: 0.0935
Epoch 785/1000
6/6 - 0s - loss: 0.0347 - mae: 0.0934
Epoch 786/1000
6/6 - 0s - loss: 0.0348 - mae: 0.0934
Epoch 787/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0932
Epoch 788/1000
6/6 - 1s - loss: 0.0346 - mae: 0.0940
Epoch 789/1000
6/6 - 0s - loss: 0.0341 - mae: 0.0930
Epoch 790/1000
6/6 - 0s - loss: 0.0339 - mae: 0.0927
Epoch 791/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0933
Epoch 792/1000
6/6 - 0s - loss: 0.0344 - mae: 0.0931
Epoch 793/1000
6/6 - 0s - loss: 0.0345 - mae: 0.0943
Epoch 794/1000
6/6 - 0s - loss: 0.0343 - mae: 0.0937
Epoch 795/1000
6/6 - 1s - loss: 0.0338 - mae: 0.0939
Epoch 796/1000
6/6 - 0s - loss: 0.0341 - mae: 0.0936
Epoch 797/1000
6/6 - 0s - loss: 0.0334 - mae: 0.0922
Epoch 798/1000
6/6 - 1s - loss: 0.0338 - mae: 0.0936
Epoch 799/1000
6/6 - 1s - loss: 0.0341 - mae: 0.0935
Epoch 800/1000
6/6 - 0s - loss: 0.0339 - mae: 0.0929
Epoch 801/1000
6/6 - 0s - loss: 0.0336 - mae: 0.0929
Epoch 802/1000
6/6 - 0s - loss: 0.0339 - mae: 0.0932
Epoch 803/1000
6/6 - 0s - loss: 0.0340 - mae: 0.0933
Epoch 804/1000
6/6 - 0s - loss: 0.0337 - mae: 0.0936
Epoch 805/1000
6/6 - 0s - loss: 0.0337 - mae: 0.0934
Epoch 806/1000
6/6 - 0s - loss: 0.0336 - mae: 0.0930
Epoch 807/1000
6/6 - 0s - loss: 0.0335 - mae: 0.0925
Epoch 808/1000
6/6 - 1s - loss: 0.0333 - mae: 0.0924
Epoch 809/1000
6/6 - 0s - loss: 0.0334 - mae: 0.0923
Epoch 810/1000
6/6 - 0s - loss: 0.0330 - mae: 0.0924
Epoch 811/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0918
Epoch 812/1000
6/6 - 0s - loss: 0.0333 - mae: 0.0930
Epoch 813/1000
6/6 - 0s - loss: 0.0331 - mae: 0.0922
Epoch 814/1000
6/6 - 0s - loss: 0.0327 - mae: 0.0922
Epoch 815/1000
6/6 - 0s - loss: 0.0330 - mae: 0.0925
Epoch 816/1000
6/6 - 0s - loss: 0.0331 - mae: 0.0927
Epoch 817/1000
6/6 - 0s - loss: 0.0330 - mae: 0.0921
Epoch 818/1000
6/6 - 0s - loss: 0.0330 - mae: 0.0924
Epoch 819/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0924
Epoch 820/1000
6/6 - 0s - loss: 0.0329 - mae: 0.0928
Epoch 821/1000
6/6 - 0s - loss: 0.0327 - mae: 0.0929
Epoch 822/1000
6/6 - 1s - loss: 0.0323 - mae: 0.0922
Epoch 823/1000
6/6 - 0s - loss: 0.0323 - mae: 0.0916
Epoch 824/1000
6/6 - 0s - loss: 0.0330 - mae: 0.0918
Epoch 825/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0919
Epoch 826/1000
6/6 - 0s - loss: 0.0327 - mae: 0.0919
Epoch 827/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0920
Epoch 828/1000
6/6 - 0s - loss: 0.0323 - mae: 0.0920
Epoch 829/1000
6/6 - 0s - loss: 0.0326 - mae: 0.0922
Epoch 830/1000
6/6 - 0s - loss: 0.0325 - mae: 0.0916
Epoch 831/1000
6/6 - 0s - loss: 0.0320 - mae: 0.0920
Epoch 832/1000
6/6 - 0s - loss: 0.0322 - mae: 0.0919
Epoch 833/1000
6/6 - 0s - loss: 0.0319 - mae: 0.0908
Epoch 834/1000
6/6 - 0s - loss: 0.0323 - mae: 0.0919
Epoch 835/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0916
Epoch 836/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0910
Epoch 837/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0915
Epoch 838/1000
6/6 - 0s - loss: 0.0322 - mae: 0.0933
Epoch 839/1000
6/6 - 1s - loss: 0.0315 - mae: 0.0917
Epoch 840/1000
6/6 - 0s - loss: 0.0316 - mae: 0.0919
Epoch 841/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0914
Epoch 842/1000
6/6 - 0s - loss: 0.0317 - mae: 0.0913
Epoch 843/1000
6/6 - 1s - loss: 0.0318 - mae: 0.0913
Epoch 844/1000
6/6 - 0s - loss: 0.0316 - mae: 0.0909
Epoch 845/1000
6/6 - 0s - loss: 0.0313 - mae: 0.0914
Epoch 846/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0906
Epoch 847/1000
6/6 - 1s - loss: 0.0314 - mae: 0.0909
Epoch 848/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0913
Epoch 849/1000
6/6 - 0s - loss: 0.0312 - mae: 0.0907
Epoch 850/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0909
Epoch 851/1000
6/6 - 0s - loss: 0.0307 - mae: 0.0900
Epoch 852/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0905
Epoch 853/1000
6/6 - 0s - loss: 0.0313 - mae: 0.0907
Epoch 854/1000
6/6 - 0s - loss: 0.0308 - mae: 0.0904
Epoch 855/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0903
Epoch 856/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0904
Epoch 857/1000
6/6 - 0s - loss: 0.0307 - mae: 0.0898
Epoch 858/1000
6/6 - 0s - loss: 0.0316 - mae: 0.0908
Epoch 859/1000
6/6 - 0s - loss: 0.0314 - mae: 0.0915
Epoch 860/1000
6/6 - 0s - loss: 0.0318 - mae: 0.0924
Epoch 861/1000
6/6 - 0s - loss: 0.0312 - mae: 0.0908
Epoch 862/1000
6/6 - 0s - loss: 0.0311 - mae: 0.0908
Epoch 863/1000
6/6 - 0s - loss: 0.0304 - mae: 0.0895
Epoch 864/1000
6/6 - 0s - loss: 0.0306 - mae: 0.0896
Epoch 865/1000
6/6 - 0s - loss: 0.0309 - mae: 0.0899
Epoch 866/1000
6/6 - 1s - loss: 0.0303 - mae: 0.0894
Epoch 867/1000
6/6 - 0s - loss: 0.0310 - mae: 0.0905
Epoch 868/1000
6/6 - 1s - loss: 0.0303 - mae: 0.0900
Epoch 869/1000
6/6 - 0s - loss: 0.0306 - mae: 0.0898
Epoch 870/1000
6/6 - 1s - loss: 0.0307 - mae: 0.0902
Epoch 871/1000
6/6 - 0s - loss: 0.0304 - mae: 0.0900
Epoch 872/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0891
Epoch 873/1000
6/6 - 0s - loss: 0.0300 - mae: 0.0893
Epoch 874/1000
6/6 - 0s - loss: 0.0305 - mae: 0.0896
Epoch 875/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0899
Epoch 876/1000
6/6 - 0s - loss: 0.0303 - mae: 0.0895
Epoch 877/1000
6/6 - 1s - loss: 0.0305 - mae: 0.0895
Epoch 878/1000
6/6 - 1s - loss: 0.0308 - mae: 0.0903
Epoch 879/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0892
Epoch 880/1000
6/6 - 0s - loss: 0.0303 - mae: 0.0893
Epoch 881/1000
6/6 - 1s - loss: 0.0302 - mae: 0.0905
Epoch 882/1000
6/6 - 1s - loss: 0.0300 - mae: 0.0890
Epoch 883/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0890
Epoch 884/1000
6/6 - 1s - loss: 0.0301 - mae: 0.0891
Epoch 885/1000
6/6 - 0s - loss: 0.0301 - mae: 0.0893
Epoch 886/1000
6/6 - 1s - loss: 0.0301 - mae: 0.0895
Epoch 887/1000
6/6 - 0s - loss: 0.0297 - mae: 0.0892
Epoch 888/1000
6/6 - 0s - loss: 0.0303 - mae: 0.0893
Epoch 889/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0896
Epoch 890/1000
6/6 - 0s - loss: 0.0297 - mae: 0.0885
Epoch 891/1000
6/6 - 0s - loss: 0.0297 - mae: 0.0887
Epoch 892/1000
6/6 - 1s - loss: 0.0299 - mae: 0.0896
Epoch 893/1000
6/6 - 1s - loss: 0.0299 - mae: 0.0899
Epoch 894/1000
6/6 - 1s - loss: 0.0298 - mae: 0.0891
Epoch 895/1000
6/6 - 1s - loss: 0.0298 - mae: 0.0890
Epoch 896/1000
6/6 - 0s - loss: 0.0300 - mae: 0.0889
Epoch 897/1000
6/6 - 1s - loss: 0.0297 - mae: 0.0889
Epoch 898/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0885
Epoch 899/1000
6/6 - 0s - loss: 0.0299 - mae: 0.0890
Epoch 900/1000
6/6 - 0s - loss: 0.0297 - mae: 0.0888
Epoch 901/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0894
Epoch 902/1000
6/6 - 0s - loss: 0.0294 - mae: 0.0898
Epoch 903/1000
6/6 - 0s - loss: 0.0293 - mae: 0.0889
Epoch 904/1000
6/6 - 0s - loss: 0.0293 - mae: 0.0883
Epoch 905/1000
6/6 - 0s - loss: 0.0293 - mae: 0.0886
Epoch 906/1000
6/6 - 0s - loss: 0.0290 - mae: 0.0882
Epoch 907/1000
6/6 - 0s - loss: 0.0292 - mae: 0.0887
Epoch 908/1000
6/6 - 0s - loss: 0.0292 - mae: 0.0880
Epoch 909/1000
6/6 - 0s - loss: 0.0291 - mae: 0.0881
Epoch 910/1000
6/6 - 0s - loss: 0.0293 - mae: 0.0888
Epoch 911/1000
6/6 - 0s - loss: 0.0293 - mae: 0.0891
Epoch 912/1000
6/6 - 0s - loss: 0.0295 - mae: 0.0887
Epoch 913/1000
6/6 - 0s - loss: 0.0293 - mae: 0.0886
Epoch 914/1000
6/6 - 0s - loss: 0.0291 - mae: 0.0882
Epoch 915/1000
6/6 - 0s - loss: 0.0289 - mae: 0.0882
Epoch 916/1000
6/6 - 1s - loss: 0.0290 - mae: 0.0887
Epoch 917/1000
6/6 - 1s - loss: 0.0291 - mae: 0.0880
Epoch 918/1000
6/6 - 0s - loss: 0.0288 - mae: 0.0879
Epoch 919/1000
6/6 - 1s - loss: 0.0284 - mae: 0.0874
Epoch 920/1000
6/6 - 0s - loss: 0.0289 - mae: 0.0877
Epoch 921/1000
6/6 - 0s - loss: 0.0289 - mae: 0.0881
Epoch 922/1000
6/6 - 1s - loss: 0.0293 - mae: 0.0884
Epoch 923/1000
6/6 - 0s - loss: 0.0297 - mae: 0.0900
Epoch 924/1000
6/6 - 0s - loss: 0.0290 - mae: 0.0898
Epoch 925/1000
6/6 - 1s - loss: 0.0286 - mae: 0.0894
Epoch 926/1000
6/6 - 0s - loss: 0.0285 - mae: 0.0882
Epoch 927/1000
6/6 - 1s - loss: 0.0287 - mae: 0.0874
Epoch 928/1000
6/6 - 0s - loss: 0.0287 - mae: 0.0878
Epoch 929/1000
6/6 - 0s - loss: 0.0286 - mae: 0.0874
Epoch 930/1000
6/6 - 1s - loss: 0.0286 - mae: 0.0874
Epoch 931/1000
6/6 - 0s - loss: 0.0287 - mae: 0.0876
Epoch 932/1000
6/6 - 1s - loss: 0.0287 - mae: 0.0878
Epoch 933/1000
6/6 - 0s - loss: 0.0283 - mae: 0.0868
Epoch 934/1000
6/6 - 0s - loss: 0.0283 - mae: 0.0869
Epoch 935/1000
6/6 - 0s - loss: 0.0285 - mae: 0.0870
Epoch 936/1000
6/6 - 1s - loss: 0.0281 - mae: 0.0867
Epoch 937/1000
6/6 - 0s - loss: 0.0283 - mae: 0.0869
Epoch 938/1000
6/6 - 1s - loss: 0.0283 - mae: 0.0873
Epoch 939/1000
6/6 - 1s - loss: 0.0279 - mae: 0.0865
Epoch 940/1000
6/6 - 0s - loss: 0.0278 - mae: 0.0861
Epoch 941/1000
6/6 - 0s - loss: 0.0282 - mae: 0.0871
Epoch 942/1000
6/6 - 1s - loss: 0.0283 - mae: 0.0870
Epoch 943/1000
6/6 - 0s - loss: 0.0279 - mae: 0.0863
Epoch 944/1000
6/6 - 1s - loss: 0.0280 - mae: 0.0865
Epoch 945/1000
6/6 - 0s - loss: 0.0283 - mae: 0.0872
Epoch 946/1000
6/6 - 1s - loss: 0.0280 - mae: 0.0865
Epoch 947/1000
6/6 - 0s - loss: 0.0284 - mae: 0.0875
Epoch 948/1000
6/6 - 0s - loss: 0.0279 - mae: 0.0872
Epoch 949/1000
6/6 - 1s - loss: 0.0280 - mae: 0.0871
Epoch 950/1000
6/6 - 0s - loss: 0.0277 - mae: 0.0866
Epoch 951/1000
6/6 - 1s - loss: 0.0275 - mae: 0.0862
Epoch 952/1000
6/6 - 0s - loss: 0.0276 - mae: 0.0866
Epoch 953/1000
6/6 - 1s - loss: 0.0280 - mae: 0.0865
Epoch 954/1000
6/6 - 0s - loss: 0.0276 - mae: 0.0860
Epoch 955/1000
6/6 - 0s - loss: 0.0273 - mae: 0.0860
Epoch 956/1000
6/6 - 0s - loss: 0.0278 - mae: 0.0863
Epoch 957/1000
6/6 - 0s - loss: 0.0278 - mae: 0.0866
Epoch 958/1000
6/6 - 0s - loss: 0.0279 - mae: 0.0868
Epoch 959/1000
6/6 - 0s - loss: 0.0274 - mae: 0.0866
Epoch 960/1000
6/6 - 0s - loss: 0.0274 - mae: 0.0860
Epoch 961/1000
6/6 - 0s - loss: 0.0273 - mae: 0.0856
Epoch 962/1000
6/6 - 1s - loss: 0.0278 - mae: 0.0866
Epoch 963/1000
6/6 - 0s - loss: 0.0272 - mae: 0.0861
Epoch 964/1000
6/6 - 0s - loss: 0.0274 - mae: 0.0859
Epoch 965/1000
6/6 - 1s - loss: 0.0277 - mae: 0.0862
Epoch 966/1000
6/6 - 0s - loss: 0.0273 - mae: 0.0861
Epoch 967/1000
6/6 - 0s - loss: 0.0275 - mae: 0.0867
Epoch 968/1000
6/6 - 0s - loss: 0.0278 - mae: 0.0862
Epoch 969/1000
6/6 - 0s - loss: 0.0268 - mae: 0.0854
Epoch 970/1000
6/6 - 1s - loss: 0.0273 - mae: 0.0856
Epoch 971/1000
6/6 - 0s - loss: 0.0276 - mae: 0.0863
Epoch 972/1000
6/6 - 0s - loss: 0.0277 - mae: 0.0862
Epoch 973/1000
6/6 - 1s - loss: 0.0271 - mae: 0.0856
Epoch 974/1000
6/6 - 0s - loss: 0.0275 - mae: 0.0865
Epoch 975/1000
6/6 - 0s - loss: 0.0272 - mae: 0.0861
Epoch 976/1000
6/6 - 0s - loss: 0.0277 - mae: 0.0862
Epoch 977/1000
6/6 - 0s - loss: 0.0267 - mae: 0.0850
Epoch 978/1000
6/6 - 0s - loss: 0.0270 - mae: 0.0853
Epoch 979/1000
6/6 - 0s - loss: 0.0272 - mae: 0.0856
Epoch 980/1000
6/6 - 0s - loss: 0.0271 - mae: 0.0857
Epoch 981/1000
6/6 - 0s - loss: 0.0270 - mae: 0.0857
Epoch 982/1000
6/6 - 1s - loss: 0.0271 - mae: 0.0858
Epoch 983/1000
6/6 - 0s - loss: 0.0275 - mae: 0.0869
Epoch 984/1000
6/6 - 1s - loss: 0.0272 - mae: 0.0868
Epoch 985/1000
6/6 - 0s - loss: 0.0271 - mae: 0.0855
Epoch 986/1000
6/6 - 0s - loss: 0.0270 - mae: 0.0856
Epoch 987/1000
6/6 - 1s - loss: 0.0270 - mae: 0.0856
Epoch 988/1000
6/6 - 0s - loss: 0.0270 - mae: 0.0856
Epoch 989/1000
6/6 - 0s - loss: 0.0271 - mae: 0.0853
Epoch 990/1000
6/6 - 0s - loss: 0.0267 - mae: 0.0857
Epoch 991/1000
6/6 - 0s - loss: 0.0263 - mae: 0.0848
Epoch 992/1000
6/6 - 0s - loss: 0.0267 - mae: 0.0846
Epoch 993/1000
6/6 - 0s - loss: 0.0264 - mae: 0.0850
Epoch 994/1000
6/6 - 0s - loss: 0.0264 - mae: 0.0849
Epoch 995/1000
6/6 - 1s - loss: 0.0267 - mae: 0.0857
Epoch 996/1000
6/6 - 0s - loss: 0.0266 - mae: 0.0847
Epoch 997/1000
6/6 - 0s - loss: 0.0264 - mae: 0.0846
Epoch 998/1000
6/6 - 0s - loss: 0.0271 - mae: 0.0851
Epoch 999/1000
6/6 - 0s - loss: 0.0266 - mae: 0.0851
Epoch 1000/1000
6/6 - 1s - loss: 0.0263 - mae: 0.0843

Fold 4 prediction cv gini_0: 0.94147


Fold 4 prediction cv gini_1: 0.93806


Fold 4 prediction cv gini_2: 0.93714


Fold 4 prediction cv gini_3: 0.93201


Fold 4 prediction cv gini_4: 0.92763


Fold 4 prediction cv gini_5: 0.92701


Fold 4 prediction cv gini_6: 0.92476

Mean validation fold gini_0: 0.93300

Mean validation fold gini_1: 0.92787

Mean validation fold gini_2: 0.92452

Mean validation fold gini_3: 0.92216

Mean validation fold gini_4: 0.91854

Mean validation fold gini_5: 0.91688

Mean validation fold gini_6: 0.91227

1.6.2 Summary and plot the LSTM model

plot_model(NN, to_file='lstm_embeddings_plot.png', show_shapes=True, show_layer_names=True)

print(NN.summary())
with open('model_summary_lstm_embeddings.txt', 'w') as f:
    with redirect_stdout(f):
        NN.summary()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-1-3111e5d9423e> in <module>()
----> 1 print(NN.summary())
      2 with open('model_summary_lstm_embeddings.txt', 'w') as f:
      3     with redirect_stdout(f):
      4         NN.summary()

NameError: name 'NN' is not defined

1.7 Organize and evaluate model performance

1.7.1 Organize and evaluate model performance: train set

# average train_preds across folds to get train_y_pred
train_y_pred_fold = np.empty((len(train_idx_fold[0]), n_steps_out, len(train_idx_fold)))
train_y_pred_fold[:]=np.NaN
for i, train_idx in enumerate(train_idx_fold): 
    train_y_pred_fold[train_idx,:,i] = train_preds[i]
train_y_pred = np.nanmean(train_y_pred_fold, axis=2)

rez_LSTM_7d_train = {}
train_set_idx = np.isnan(train_y_pred).sum(axis=1)==0
rez_LSTM_7d_train['timestamp'] = [train_timestamp[i] for i, logic in enumerate(train_set_idx) if logic ==True]
rez_LSTM_7d_train['country'] = [train_country[i] for i, logic in enumerate(train_set_idx) if logic ==True]
rez_LSTM_7d_train['X'] = train_X[train_set_idx,:] # note that variables are scaled and preprocessed for embedding 
rez_LSTM_7d_train['y'] = scaler_y.inverse_transform(train_y)[train_set_idx,:] # back to the original unit (cases per million)
rez_LSTM_7d_train['y_pred'] = scaler_y.inverse_transform(train_y_pred)[train_set_idx,:] 
rez_LSTM_7d_train['rmse'] = rmse_y_y_pred(rez_LSTM_7d_train, n_steps_out)

1.7.2 Organize and evaluate model performance: validation set

# average val_preds across folds to get val_y_pred
val_y_pred_fold = np.empty((len(train_idx_fold[0]), n_steps_out, len(train_idx_fold)))
val_y_pred_fold[:]=np.NaN
for i, val_idx in enumerate(val_idx_fold): 
    val_y_pred_fold[val_idx,:,i] = val_preds[i]
val_y_pred = np.nanmean(val_y_pred_fold, axis=2)

rez_LSTM_7d_val = {}
val_set_idx = np.isnan(val_y_pred).sum(axis=1)==0
rez_LSTM_7d_val['timestamp'] = [train_timestamp[i] for i, logic in enumerate(val_set_idx) if logic ==True]
rez_LSTM_7d_val['country'] = [train_country[i] for i, logic in enumerate(val_set_idx) if logic ==True]
rez_LSTM_7d_val['X'] = train_X[val_set_idx,:] # note that variables are scaled and preprocessed for embedding 
rez_LSTM_7d_val['y'] = scaler_y.inverse_transform(train_y)[val_set_idx,:] # back to the original unit (cases per million)
rez_LSTM_7d_val['y_pred'] = scaler_y.inverse_transform(val_y_pred)[val_set_idx,:] 
rez_LSTM_7d_val['rmse'] = rmse_y_y_pred(rez_LSTM_7d_val, n_steps_out)

1.7.3 Organize and evaluate model performance: test set

rez_LSTM_7d_test = {}
rez_LSTM_7d_test['timestamp'] = test_timestamp
rez_LSTM_7d_test['country'] = test_country
rez_LSTM_7d_test['X'] = test_X # note that variables are scaled and preprocessed for embedding 
rez_LSTM_7d_test['y'] = scaler_y.inverse_transform(test_y)
rez_LSTM_7d_test['y_pred'] = scaler_y.inverse_transform(test_y_pred)
rez_LSTM_7d_test['rmse'] = rmse_y_y_pred(rez_LSTM_7d_test, n_steps_out)

1.7.4 Organize and evaluate model performance: save data

rez_LSTM_7d = {'train':rez_LSTM_7d_train, 'validation':rez_LSTM_7d_val, 'test':rez_LSTM_7d_test}

filePath_pickle = Path('/content/drive/My Drive/Colab_data/covid19/covid19_LSTM_7d_with_lagging_shorter_test_period_colab.pickle')
with open(filePath_pickle, 'wb') as f:
    pickle.dump(rez_LSTM_7d, f)

#load the saved dictionary from pickle file
#filePath_pickle = Path('/content/drive/My Drive/Colab_data/covid19/covid19_LSTM_7d_with_lagging_shorter_test_period_colab.pickle')
#with open(filePath_pickle, 'rb') as f:
#     rez_LSTM_7d = pickle.load(f)

1.7.5 Organize and evaluate model performance: plot

plot_actual_predicted(rez_LSTM_7d['train'], dict_country, 'train', n_steps_out=7)

1.7.6 Analyze tensorboard data

# upload tensorboard data to dev to make it publicly available
!tensorboard dev upload --logdir ./logs \
  --name "multivariate_timeseries_energy_usage" \
  --description "Training results from /Volumes/GoogleDrive/My Drive/Colab Notebooks/multivariate_timeseries_analysis_energy_usage_.ipynb" \
  --one_shot

# if the above upload is successful, the specific experiment sould be listed using the command below
!tensorboard dev list

# the following describes how to get the tensorboard experiment data as pandas dataframe using the api (https://www.tensorflow.org/tensorboard/dataframe_api)
import tensorboard as tb
experiment_id = "c1KCv3X3QvGwaXfgX1c4tg" # get the proper experiment id
experiment = tb.data.experimental.ExperimentFromDev(experiment_id)
df = experiment.get_scalars()